/* ---- Google Analytics Code Below */

Friday, March 31, 2023

ChatGPT banned in Italy over privacy, data collection concerns

Unexpected Privacy concerns posed.

Calls have grown in the United States to stop development of the AI technology    By Kelsey Koberg

DataGrade founder Joe Toscano and MRC Free Speech America VP Dan Schneider said ChatGPT poses various dangers to jobs and information.   

Experts say biased data in ChatGPT could make AI ‘more dangerous,’ impact journalism

Italy’s privacy regulator ordered a ban Friday on ChatGPT over alleged privacy violations.  ... '

Reading Minds with AI: Researchers Translate Brain Waves to Images

Quite a considerable capability suggested.   in KDNuggets

Reading Minds with AI: Researchers Translate Brain Waves to Images

Two researchers from Osaka University were able to reconstruct highly accurate images from human brain activity obtained by fMRI. Read this article if you are curious to find out what all the hype is about.

By Kanwal Mehreen, Software Engineering Student at NUST on March 28, 2023 in Artificial Intelligence    

Microsoft to Ruin Chat Search?

Ouch if this is the case. 

ChatGPT Opened a New Era in Search. Microsoft Could Ruin It

By Wired, March 31, 2023

Microsoft in February announced its standard fees for search data would increase by as much as 10 times starting in May.

Google typically gets the blame for the lack of competition in web search. The US government is even suing to block the company from using allegedly monopolistic tactics, like making itself the default search engine in widely used software such as Android, Chrome, and Safari.

But some upstart search engines trying to woo users with privacy protections or ad-free searches say their latest challenge doesn't come from Google. Instead, it's Microsoft and its Bing search engine causing their aggravation.

Search startups have long relied on licensing search results from Bing, tapping a web indexing operation larger than a small company could easily afford and adding their own features and ways of parsing queries. But Microsoft's rollout of a Bing search chatbot based on technology underlying OpenAI's ChatGPT has prompted concerns that Microsoft is unfairly squeezing out its search data customers as it launches a renewed attempt to bite off more market share from Google.

From Wired

View Full Article  

Should AI Insights be Shared with Consumers, and How?

 Interesting thoughts.    Are analytic marketing methods shared with consumers?    Resources at link. 

Should AI Insights Be Shared With Consumers?   by Tom Ryan  in Retailwire   plus expert opinion.

Tomas Chamorro-Premuzic, a psychology professor at University College London, argues In a column for Harvard Business Review that while artificial intelligence (AI) technology is helping brands understand their customers better, it can also help customers understand themselves better.

“Your choice of music reveals the degree to which you are extraverted, curious, and neurotic; your choice of movies reveals the degree to which you are intelligent, conscientious, and agreeable; your Facebook data reveals whether you are conservative or liberal, sociable or introverted, optimistic or pessimistic; your tweets reveal whether you are narcissistic or not, and so on,” he writes.

Prof. Chamorro-Premuzic posits that AI could be used to discover, interpret and share these patterns. The algorithmic insights, he adds, can offer “granular and personalized” reads into preferences and personality traits in real time to help individuals “become more self-aware.”

“Just as wearables can translate physiological signals into actionable feedback on our fitness, energy, sleepiness, or stress levels, AI could detect changing patterns to our habits to alert us about increases in negative or positive affect, curiosity, or aggression,” he writes.

AI’s potential for therapy – at least as a complementary tool – is being increasingly examined, particularly given the financial and logistical barriers facing human-centered treatment.

In the UK, an AI chatbot, Limbic Access, was recently approved for medical use. It promises to predict mental health disorders with 93 percent accuracy.

A recent NPR article stated, “Advances in artificial intelligence — such as ChatGPT — are increasingly being looked to as a way to help screen for, or support, people who are dealing with isolation, or mild depression or anxiety. Human emotions are tracked, analyzed and responded to, using machine learning that tries to monitor a patient’s mood, or mimic a human therapist’s interactions with a patient.”

Prof. Chamorro-Premuzic said brands sharing AI-driven insights into their consumers’ personalities could help overcome the “creepy” feeling that comes with personalized targeting. He wrote, “Brands will enhance their ethical reputation and trustworthiness if they share this understanding with consumers, persuading them that there is no conflict between knowing them well, and helping them know themselves well, when done in an ethical and transparent way.”

Automakers Develop Robot That Charges Your EV for You

Essential?  But many things are not  

Automakers Develop Robot That Charges Your EV for You

For the EV fence-sitters who find public charging cables too heavy or confusing, Hyundai and Kia have developed a potential solution: the Automatic Charging Robot.

By Adrianna Nine March 27, 2023

As automakers and climate-focused legislators continue to push electric vehicle adoption, they’ll need all the help they can get. If the idea of using public EV charging cables gives you pause, Hyundai and Kia have developed the Automatic Charging Robot. 

Shown off online last week, Hyundai Motor Group’s ACR aims to make EV charging more convenient by doing the heavy lifting for you. The robot consists of a sleek safety pole, which sits atop a wheeled base. An arm that holds the charging cable at one point and the charging plug at another is attached to the pole. Drivers park in a designated charging spot (or tell their car to do that for them) and hit a button on the key fob that tells the ACR to initiate charging. 

ACR sends a signal that commands the car to open its charging port door. Once that’s open, the ACR wheels closer to the vehicle and uses its AI-equipped 3D camera to line up the charging plug with the port. A meter on the side of the ACR’s safety pole displays the vehicle’s current charge level. Once that meter hits 100%, the robot arm pulls back, and the ACR wheels away.  ...' 


Thursday, March 30, 2023

What Early Adopters Can Teach Us About AI

 Some good thoughts.  

What Early Adopters Can Teach Us About AI

Interview with Thomas H. Davenport   in APQC

ChatGPT burst upon the AI landscape in November 2022 with a media and market frenzy not seen since Steve Jobs introduced the iPhone in 2007. Over one million people immediately signed up to test if OpenAI’s large language learning algorithm could perform as well or better than a human at answering text-based research questions, writing a poem in the style of Shakespeare or songs like Bob Dylan, completing high school homework assignments (you can imagine the handwringing here), and much more. In many cases, the answer was yes.  

Only three months later, there have already been three major upgrades to the search and auto correct algorithm. ChatGPT4 gets the answers wrong and “hallucinates” (makes up stuff) less frequently. Competitors from Google and many others have joined the race to use AI to empower everything we do online and virtually every process we use inside organizations.

As we all know, the promise of AI won’t be actualized unless we develop a strategy, explore use cases, and operationalize the technology in our organizations. This takes organizational savvy and change management skills, which no algorithm can give you. 

Fortunately, we can learn from multifaceted experts like Tom Davenport and a decade of early adopters who have used AI to dramatically accelerate their businesses. In addition to dozens of books and articles on topics like knowledge management and analytics, Tom is co-author (with Nitin Mittal) of All in On AI: How Smart Companies Win Big with Artificial Intelligence. In this interview excerpt, Tom speaks with me about what it means to be “All in” on AI, shares some key lessons from early adopters, and provides insight into what AI will mean for the workplaces of the future.  

What does it mean to be “all in” on AI?

Organizations that are “all in” on AI are highly invested in a variety of different ways. One is that they have different types of AI spread throughout the company quite broadly. We’re talking about dozens of use cases at a minimum, but more commonly it’s hundreds or even thousands of use cases because AI is a narrow technology. It tends to support or automate tasks, not entire jobs—and certainly not entire processes. So for example, if you want to automate all customer service or all order management, you will need to assemble many different pieces of AI to have a high level of impact. 

Being all in means you’re not just using AI for optimization at the margins of your business but really changing something dramatically. In the book, we talk in terms of organizations that use AI to transform their strategy, business model, products, services, operations, or even customer behaviors—not just making small tweaks for operational improvement. 

You’re probably going to get into trouble unless you’re ethical about the ways in which you use AI. Organizations that are all-in also have a framework for ethical and trustworthy AI in place that includes guidelines, policy approaches, and governance structures. 

What led you to focus on early adopters of AI in your book?

I wrote a book called Competing on Analytics in 2007. My readers found it quite helpful to look at companies that were aggressive early adopters of analytics, and I thought that the same thinking probably applies with regard to AI. The people I was working with at Deloitte were starting to talk about the ways in which AI could help transform these big legacy companies, a lot of whom were their clients. I think the nice thing about those examples is that even if people can’t or don’t want to go all in, they respond to reading about what it’s like to be all in and what companies can accomplish if they’re really aggressive in their adoption of the technology.  ... ' 

What Training/Testing Data?

Interesting, was wondering which training data was used where, and the consequences.  Is that readily testable?  

ARTIFICIAL INTELLIGENCE/TECH/GOOGLE

Google denies Bard was trained with ChatGPT data

The Information published a report Wednesday including allegations from a former Google AI researcher that the company used a rival’s responses to train its own chatbot. Google denies that Bard uses that data.

By SEAN HOLLISTER

Mar 29, 2023, 10:10 PM EDT|19 Comments / 19 New

DeepMind reportedly lost a yearslong bid to win more independence from Google

OpenAI co-founder on company’s past approach to openly sharing research: ‘We were wrong’

AI chatbots compared: Bard vs. Bing vs. ChatGPT

Google’s Bard hasn’t exactly had an impressive debut — and The Information is reporting that the company is so interested in changing the fortunes of its AI chatbots, it’s forcing its DeepMind division to help the Google Brain team beat OpenAI with a new initiative called Gemini. The Information’s report also contains the potentially staggering thirdhand allegation that Google stooped so low as to train Bard using data from OpenAI’s ChatGPT, scraped from a website called ShareGPT. A former Google AI researcher reportedly spoke out against using that data, according to the publication.

But Google is firmly and clearly denying the data was used: “Bard is not trained on any data from ShareGPT or ChatGPT,” spokesperson Chris Pappas tells The Verge.  ... ' 

Unexpected Lunar Water Discovered

Easier colonization possible?

Lunar Water Reservoir Discovered in Glass Impact Beads

Tiny glass beads scattered across the Moon may contain enough water to sustain future crewed missions.

By Ryan Whitwam March 28, 2023No humans have set foot on the Moon for more than 50 years, but the coming decade could see multiple crewed landings and the construction of an orbiting lunar station. Harvesting water from the Moon could make a permanent human presence easier to maintain, but where will we find it? A new study from the Chinese Chang’e rover team suggests there might be usable water everywhere on the Moon.

In the past, scientists believed the Moon was just a big ball of dust, but robotic missions eventually confirmed water ice exists in shadowy craters. Water also appears to evaporate from lunar soil when exposed to sunlight. New data from the Chang’e 5 mission reveals the details.

Chang’e 5, which landed on the Moon in 2020, was China's first lunar sample return mission. It sent back 3.7 pounds (1.7 kilograms) of lunar regolith, some collected from a meter below the surface. Throughout the samples, the researchers from the Chinese Academy of Sciences spotted glass beads resulting from meteorite impacts. The researchers analyzed the chemical composition and structure of the impact glass, finding they can be up to 0.2% water by mass. In absolute terms, the researchers calculate that lunar soil has between 300 billion and 270 trillion kilograms of water locked up in glass beads. If collected in one place, it could fill Lake Tahoe almost twice over at the high end of the estimate.  ... ' 

Early Looks at Ernie

 No direct testing seen yet.

Baidu Unveils Generative AI Ernie Bot With Pre-Recorded Videos, Disappointing Investors,   

 Eric Hal Schwartz in Voicebot.ai

Baidu introduced its generative Ernie Bot in a series of video presentations that promised a lot of features without actually demonstrating them live. The scripted nature of the event led to a brief dip of its Hong Kong shares by nearly 10%, though planning ahead did ensure Baidu avoided a gaffe like the incorrect answer to a question produced by the Google Bard generative AI chatbot when it was first revealed last month.

BAIDU ERNIE BOT

Baidu CEO Robin Li introduced Ernie, which stands for Enhanced Representation through Knowledge Integration, at the event. He played several videos showing Ernie performing math, generating images and videos from text prompts, and speaking in multiple Chinese dialects. The AI performed flawlessly in the video, but investors were apparently eager to see how it would perform in the wild, leading to the stock dip. Ernie is not widely available, with only those who received invitation codes from Baidu able to interact with the bot. Companies interested in employing Ernie for enterprise purposes must apply for the chance to link their software to Baidu’s cloud platform hosting the AI model.

Li admitted that Ernie is still not perfect, which might be why the company went for videos over live demonstrations. Li said Ernie had to be released “because the market demands it.” That seems to be backed up if Baidu’s boast of 650 companies in the Ernie ecosystem and 30,000 corporate user applications are to be believed.

Despite investor disappointment, Baidu has been working toward conversational generative AI for some time. The company has engaged in AI research for all kinds of products, including a virtual human named Xijiajia capable of painting a picture of a cat upon request. Xijiajia is Baidu’s digital human ambassador, one of several such virtual beings produced by Baidu, including the “interactive virtual idol” DuXiaoxiao. Some of the demand may also be a result of restrictions on American generative AI chatbots. Chinese social media platform WeChat banned ChatGPT last year, and strict rules on deepfakes have set up a potentially tricky regulatory system for any AI that can mimic specific people’s way of talking.

“Baidu envisions a future where we join forces with all to drive the evolution of AI, empowering every individual with access to state-of-the-art productivity tools and ensuring that the benefits of these advancements are shared by all,” Li said in a statement.  ... 

Comm Network for the Moon

Spreading Space Networks.

Lockheed Spinoff Plans to Build Communication Network for the Moon

Crescent Space will begin deploying its Parsec satellites in orbit on the Moon in 2025.

By Ryan Whitwam March 30, 2023

Humankind has been relegated to low-Earth orbit for decades, but that will change with the Artemis Program. NASA and its partners plan to establish a permanent human lunar presence, which will require infrastructure. Lockheed Martin has announced the formation of a new company called Crescent Space to start building it. Its goal is to deploy a satellite constellation in orbit of the Moon to provide communication and navigation services.

The network will be called Parsec, consisting of an indeterminate number of nodes built on established Lockheed Martin tech. Parsec will use Lockheed's Curio SmallSat platform, along with the SmartSat software and COMPASS/Horizon mission planning system. By placing enough of these compact satellites in orbit, it will be possible to maintain contact with all missions orbiting the Moon and any assets on the surface. As a bonus, the network will be able to offer accurate lunar positioning in the same way the GPS network does on Earth. So, even as the Moon becomes more congested, spacecraft and surface equipment can steer clear of other missions.  ... ' 

Wednesday, March 29, 2023

A Call to Pause 'Giant AI Experiments'

 Shall we pause, will every one?  For 6 months.  Dangerous?  What difference will it make? 

ARTIFICIAL INTELLIGENCE/TECH

Elon Musk and top AI researchers call for pause on ‘giant AI experiments’

 An open letter says the current race dynamic in AI is dangerous, and calls for the creation of independent regulators to ensure future systems are safe to deploy.

By JAMES VINCENT  in The Verge

Mar 29, 2023, 5:08 AM EDT|96 Comments / 96 New

A number of well-known AI researchers — and Elon Musk — have signed an open letter  calling on AI labs around the world to pause development of large-scale AI systems, citing fears over the “profound risks to society and humanity” they claim this software poses.

The letter, published by the nonprofit Future of Life Institute, notes that AI labs are currently locked in an “out-of-control race” to develop and deploy machine learning systems “that no one — not even their creators — can understand, predict, or reliably control.”

“We call on all AI labs to immediately pause for at least 6 months the training of AI systems more powerful than GPT-4.”

“Therefore, we call on all AI labs to immediately pause for at least 6 months the training of AI systems more powerful than GPT-4,” says the letter. “This pause should be public and verifiable, and include all key actors. If such a pause cannot be enacted quickly, governments should step in and institute a moratorium.”

Signatories include author Yuval Noah Harari, Apple co-founder Steve Wozniak, Skype co-founder Jaan Tallinn, politician Andrew Yang, and a number of well-known AI researchers and CEOs, including Stuart Russell, Yoshua Bengio, Gary Marcus, and Emad Mostaque. The full list of signatories can be seen here, though new names should be treated with caution as there are reports of names being added to the list as a joke (e.g. OpenAI CEO Sam Altman, an individual who is partly responsible for the current race dynamic in AI).

The letter is unlikely to have any effect on the current climate in AI research, which has seen tech companies like Google and Microsoft rush to deploy new products, often sidelining previously-avowed concerns over safety and ethics. But it is a sign of the growing opposition to this “ship it now and fix it later” approach; an opposition that could potentially make its way into the political domain for consideration by actual legislators.

As noted in the letter, even OpenAI itself has expressed the potential need for “independent review” of future AI systems to ensure they meet safety standards. The signatories say that this time has now come.

“AI labs and independent experts should use this pause to jointly develop and implement a set of shared safety protocols for advanced AI design and development that are rigorously audited and overseen by independent outside experts,” they write. “These protocols should ensure that systems adhering to them are safe beyond a reasonable doubt.”

Industrial Metaverse

 A transparent collection of interacting digital Twins?

HUMANS AND TECHNOLOGY

The emergent industrial metaverse

An interface between the real and digital worlds will transform how we work, live, and interact.

By MIT Technology Review Insightsarchive page, March 29, 2023

In partnership withSiemens

The industrial metaverse—a metaverse sector that mirrors and simulates real machines, factories, cities, transportation networks, and other highly complex systems—will offer to its participants fully immersive, real-time, interactive, persistent, and synchronous representations and simulations of the real world. 

The emergent industrial metaverse

Download the report

Existing and developing technologies, including digital twins, artificial intelligence and machine learning, extended reality, blockchain, and cloud and edge computing, will be the building blocks of the industrial metaverse. These will converge to create a powerful interface between the real and digital worlds that is greater than the sum of its individual parts. 

Annika Hauptvogel, head of technology and innovation management at Siemens, describes the industrial metaverse as “immersive, making users feel as if they’re in a real environment; collaborative in real time; open enough for different applications to seamlessly interact; and trusted by the individuals and businesses that participate”—far more than simply a digital world. 

The industrial metaverse will revolutionize the way work is done, but it will also unlock significant new value for business and societies. By allowing businesses to model, prototype, and test dozens, hundreds, or millions of design iterations in real time and in an immersive, physics-based environment before committing physical and human resources to a project, industrial metaverse tools will usher in a new era of solving real-world problems digitally.   ... ' 


Twitter is Dying?

Doubt it, I still get an overload of tweets, mostly tech and not politics  Get the impression that many still live there. 

Twitter is Dying?

Natasha Lomas@riptari / 8:20 AM EDT•March 28, 2023

It’s five months since Elon Musk overpaid for a relatively small microblogging platform called Twitter. The platform had punched above its weight in pure user numbers thanks to an unrivaled ability to both distribute real-time information and make expertise available. Combine these elements with your own critical faculty — to weed out the usual spam and bs — and it could feel like the only place online that really mattered.

Even if the average internet user remained baffled by Twitter, it contained essential ingredients that made it a go-to source for journalists or other curious types wanting to earwig on conversations between interesting people — whether subject experts or celebrities. It was also therefore a place where experts and celebrities could find community and an engaged audience — without the need for layers of message-filtering middlemen. Twitter was where these two sides met and (sometimes) meshed in messy conversation.  ..... ' 

Teslas Being Hacked for $$

A well known kind of security breach worked on a Tesla.  Big finding reward.  Expected to be remotely fixed.

Pwn2Own Hackers Breach a Tesla Twice,   By PC Magazine, March 29, 2023

Tesla’s security response team validated the results. The automaker is expected to issue over-the-air fixes to patch the flaws, according to SecurityWeek.

Participants of the Pwn2Own software exploitation conference hacked technology from automaker Tesla twice at the Zero Day Initiative's Pwn2Own software exploitation conference, earning $350,000 and a Model 3 infotainment system.

The team from French security company Synacktiv executed a time-of-check-to-time-of-use (TOCTOU) exploit against a Tesla Gateway, then employed a heap overflow and an out-of-band write vulnerability to gain access to and compromise the Model 3.

Pwn2Own describes a TOCTOU exploit as a "file-based race condition that occurs when a resource is checked for a particular value, and that value changes before the resource is used, invalidating the results of the check."

SecurityWeek said Tesla is expected to release patches to correct the flaws exposed by the Synacktiv hacks.

From PC Magazine  

More Details, but Not Enough

Interesting point, but do not fully understand.    Posting to be revisited.

More Details, but Not Enough

By Gregory Goth, Commissioned by CACM Staff, March 29, 2023

Nearly two years since the publication of the paper in Nature, Google has not yet fully open-sourced the data or code on which its claims were based.

The contentious discussion over the validity of Google researchers' claim that machine learning agents could achieve superhuman results in creating plans for computer chips entered a new, more public phase Tuesday (March 28), with a leading researcher in design automation finding the Google technology did not perform as its authors claimed in a paper published nearly two years ago in Nature.

The dispute around the Nature paper's claims has bubbled for nearly a year in prepared public statements and GitHub code repositories and FAQ sections; researchers directly involved in the situation have declined to speak extemporaneously for the public record. Even some subject matter experts have not wished to speak openly, given Google's dominant position in its ability to distribute research resources to academic computer scientists. However, Tuesday's presentation by Andrew Kahng, a prominent University of California, San Diego researcher in the field of electronic design automation (EDA), at the 2023 ACM/IEEE International Symposium on Physical Design, could elevate the issue to a more open avenue of argument among industry and academic experts.

Briefly stated, the authors of the Nature paper claimed their reinforcement learning (RL) agents could revolutionize the labor-intensive task of floorplanning—the architecting of the incredibly intricate network of memory components (called macro blocks) and logic circuitry (standard cells) on a chip. "Our method generates manufacturable chip floorplans in under six hours, compared to the strongest baseline, which requires months of intense effort by human experts," the authors wrote.

Kahng served as a peer reviewer for the paper, and also wrote an encapsulation for the news and views section of the journal, quoting science fiction author Arthur C. Clarke's observation that any sufficiently advanced technology is indistinguishable from magic.

"To long-time practitioners in the fields of chip design and design automation, (lead author Azalia) Mirhoseini and colleagues' results can indeed seem magical," Kahng wrote.

How open is open?

Science is not magic, however, and the Google paper's claims took the research community by storm. At the conclusion of his summation, Kahng wrote, "We can therefore expect the semiconductor industry to redouble its interest in replicating the authors' work, and to pursue a host of similar applications throughout the chip-design process."

For researchers who presumably were interested in trying to replicate those results, the Google team noted at the end of the paper that "the data supporting the findings of this study are available within the paper and the Extended Data," and that "the code used to generate these data is available from the corresponding authors upon reasonable request."

Zoom IQ Updates

 Zoom IQ Does some Interesting Tool things.  Note taking interesting. 

Zoom IQ Assistant Adds Generative AI Tools Fueled by Multiple LLMs

Eric Hal Schwartz on March 27, 2023 at 3:00 pm

Zoom has unveiled a new suite of generative AI tools for its Zoom IQ digital assistant. The “smart companion” uses a mix of large language models built in-house and by partners like OpenAI to help draft communications and condense meeting transcriptions.

ZOOM IQ

Zoom IQ launched a year ago as a meeting transcription feature capable of collecting and analyzing data from customer interactions. The upgraded AI Assistant incorporates several generative AI text features on top of that function. The new Zoom IQ meeting summary feature doe as its name suggests and condenses a meeting transcript summarizing the conversation and highlighting any next steps decided during the meeting. The summary is then shared with anyone that the meeting attendees choose.

The Zoom IQ chat compose feature provides suggestions for messages to send in Zoom Team Chats based on the context of the conversation, with different tones available. The Zoom IQ email compose feature performs a similar task but funnels previous Zoom meetings, phone calls, and email threads to formulate draft emails and is only available with Zoom IQ for Sales. Zoom IQ relies on different LLMs depending on the user’s request. Some Zoom customers will be able to augment the mix provided by Zoom and OpenAI with their own models. The idea is to up the AI’s flexibility by widening the model options, as well as allowing customers to customize the models for a company’s vocabulary and scenarios.

“Zoom has long built AI solutions into our products to empower customers to be more productive,” Zoom chief product officer Smita Hashim said. “We are excited to bring many more capabilities with new large language models. Our unique approach to AI will give customers the flexibility they want and help significantly improve collaboration and customer relations.”

Zoom’s pivot to generative AI fits the rush among enterprise service providers to incorporate the technology in their platforms. That includes investing in startups like Supernormal, which raised $10 million earlier this year while incorporating OpenAI’s models into its meeting AI assistant. Otter.ai created its own AI assistant named OtterPilot to produce summaries and notes from meetings, including slides and other images. Meanwhile, OpenAI’s majority investor Microsoft now employs ChatGPT as a note-taker for Teams.  ... ' 

Disney Scraps Metaverse?

 It would seem a natural thing  for Disney to experiment seriously with Metaverses.

Disney reportedly scraps its metaverse unit

Disney appears to be the latest company to pull back the metaverse, as CEO Bob Iger reverses more decisions from his predecessor.

By Chris Morris in FastCompany

Disney has a long history of creating fantastical worlds, but even the Mouse House now seems to be pulling back from the metaverse.

The entertainment giant’s metaverse team has reportedly been eliminated as part of the ongoing round of job cuts at Disney. The Wall Street Journal reports all 50 team members who were developing the company’s metaverse strategies have been let go.

The metaverse was a pet project of former CEO Bob Chapek, who called it “the next great storytelling frontier.”

Last February, he told employees the goal of the metaverse unit was to “create an entirely new paradigm for how audiences experience and engage with our stories.” Among his ideas were a virtual Haunted Mansion ride where people could get off the ride and explore some of the special effects. He also envisioned a world where people who went to the physical parks, if they gave permission, could have their Disney+ experience programmed, not according to what they watched last or the habits of other views, but on what they experienced while at the park in an Amazon Prime-like program.

That specific idea, the Journal reports, has been abandoned.

And as early as 2021, Josh D’Amaro, chairman of parks, experiences, and products at Disney, had mentioned bringing a metaverse to Disney World’s Main Street, which would combine the park’s entryway with digital assets, ideally creating a deeper storytelling experience.

“We’re proud of our legacy,” he said at the time. “But at the same time, we have an obligation to defy that—and push our theme park experience into the future. What we’ve shown ourselves in the past year [is] we can flex. We can be pretty quick and pretty bold in our decision-making.”

Last November, however, Chapek was unceremoniously dismissed from Disney. And Bob Iger returned to the CEO role, quickly reversing several of Chapek’s decisions.

The metaverse ambitions appear to be the latest to be targeted by Iger. Despite the many ideas that were publicly floated, the metaverse unit had apparently not made much progress on the new technology and did not have many clear plans in place.

Mike White, a former Disney consumer-products executive who headed the metaverse team, is reportedly still with the company.  ... '

Tuesday, March 28, 2023

IBM Watson StudioAI at Masters?

Below possibly methods being used for Masters demonstration.  I had previously looked at Chef Watson.

From OpenAI GPT:

IBM Watson Studio AutoAI is a cloud-based tool that helps users to automatically build, train and deploy machine learning models. It uses a combination of techniques, including deep learning and generative adversarial networks (GANs), to generate synthetic data sets that can be used for training AI models.

The GANs are used to generate synthetic data by creating two neural networks: a generator network and a discriminator network. The generator network generates synthetic data based on a given set of inputs, while the discriminator network evaluates the synthetic data and compares it with the real data set. By iterating through this process, the generator network learns to create more realistic synthetic data sets that can be used for training AI models.

In addition to using GANs for generating synthetic data sets, IBM has also worked on developing generative AI models for other applications. For example, IBM Research has developed a system called "AI Composer" that uses deep learning to generate original musical compositions. They have also developed a system called "Chef Watson" that uses machine learning to generate new and innovative recipe ideas.

IBM Plans Predictive and Generative Masters Tournament Experience

 Should be an interesting demonstration by IBM.   Hint at some predictive analytics?  Will watch and report on this.  

IBM Brings Generative AI Commentary and Hole-by-Hole Player Predictions to the Masters Digital Experience

IBM unveils new iteration of its "Let's create" campaign during this year's Masters Tournament, featuring golfer Kurt Kitayama

Mar 28, 2023

ARMONK, N.Y., March 28, 2023 /PRNewswire/ -- IBM (NYSE: IBM) and the Masters Tournament, today introduced two innovative new features as part of the award-winning Masters app and Masters.com digital experience, including Artificial Intelligence (AI) generated spoken commentary. Expanding on the popular MyGroup feature — which enables patrons of the Masters digital platforms to watch every shot, on every hole, from all their favorite players — the AI commentary solution will produce detailed golf narration for more than 20,000 video clips over the course of the Tournament. It is the latest example of how IBM and the Masters are working together to create digital fan experiences that offer unparalleled access and in-depth insights into every moment of the Tournament, from the first drive on the first tee to the final putt on the 18th green. 

Experts from IBM iX, the experience design partner within IBM Consulting, worked with the Masters digital team to leverage multiple foundation models — including IBM Watson Text-to-Speech – to train the AI in the unique language of golf at the Masters, automating the process of adding spoken commentary to video clips. Generative AI built on foundation models was applied to produce narration with varied sentence structure and vocabulary, avoiding redundant play-by-play iterations to make the clips informative and engaging.

Also new to the Tournament this year, IBM will introduce hole-by-hole player predictions. To project a player's score on each hole for the entire Tournament, the IBM Consulting team leveraged AutoAI capabilities in IBM Watson Studio to train AI models using six years of Masters data — more than 120,000 golf shots. After the completion of a given hole, the hole-by-hole projections are updated to reflect the most recent performance of the player. The new solution expands on the predictive intelligence of the popular Players Insights and Masters Fantasy Projections feature, which turns data into insights around the most exciting holes to watch for every golfer, the low/high score for all golfers, and projected Masters Fantasy points for every round.

"For more than twenty years, IBM and the Masters have partnered together to create custom solutions that improve the user experience and capture the attention of millions of golf fans globally," said Jonathan Adashek, Senior Vice President of Marketing and Communications for IBM. "We're applying the same hybrid cloud, AI technology and IBM Consulting services that we use with clients across industries to bring an elevated digital experience to one of the most iconic sporting events in the world."

Timed to this year's Tournament, IBM will launch the next iteration of its "Let's create" brand initiative, "What if?". Conceptualized and executed by Ogilvy, "What if?" underscores the importance of asking big questions that can lead to the creation of equally big technology solutions. The spots were filmed by ProdCo's Ian Pons Jewell, featuring Golden Globe-winning and Emmy-nominated actor and producer Oscar Isaac as the voice over interacting with people in each scene. One installment of the four distinct ads includes rising golfer and 87th Masters Tournament invitee Kurt Kitayama and sports commentator Jim Nantz, who bring this creative concept to life through the lens of IBM's collaborative partnership with the Masters. In addition to the Masters, the campaign is inspired by IBM's work with a range of clients across sustainability, supply chain, and data analytics and security, demonstrating how IBM is helping organizations drive innovation, transformation, and enhanced customer experiences. The campaign begins April 5 across broadcast, connected TV, digital video, print and social.  ... ' 

NFT Plan Dropped

From the BBC, indicator of a bubble? 

NFT: Plans for Royal Mint produced token dropped

Plans for the NFT for Britain were launched when Rishi Sunak was chancellor

By Helen Catt & Sam Francis

Plans for a government backed non-fungible token (NFT) produced by the Royal Mint have been dropped, the Treasury has announced.

Rishi Sunak ordered the creation of a "NFT for Britain" that could be traded online, while chancellor in April 2022.

NFTs are assets in the digital world that can be bought and sold, but which have no physical form of their own.

The Royal Mint announced it was "not proceeding with the launch" following a consultation with the Treasury.

Economic Secretary Andrew Griffiths said the department would keep the proposal "under review".

NFTs have been touted as the digital answer to collectables, but some sceptics fear they could be a bubble waiting to burst. They have been used as speculative assets and some have sold for millions of dollars.

Responding to the announcement, Harriet Baldwin, chair of the Treasury Select Committee, said: "We have not yet seen a lot of evidence that our constituents should be putting their money in these speculative tokens unless they are prepared to lose all their money.

"So perhaps that is why the Royal Mint has made this decision in conjunction with the Treasury."  ... ' 

Amazon To Go Head to Head with SpaceX

 Satellite Internet Dominance

ACM NEWS

Amazon To Go Head to Head with SpaceX in a Battle for Satellite Internet Dominance

By MIT Technology Review, March 27, 2023

The rocket's inaugural flight will carry two prototype Project Kuiper satellites to orbit.

A booster for the United Launch Alliance's new heavy-lift Vulcan Centaur rocket is readied for integration.

Credit: United Launch Alliance

Elon Musk and Jeff Bezos are about to lock horns once again. Last month, the U.S. Federal Communications Commission (FCC)  approved the final aspects of Project Kuiper, Amazon's effort to deliver high-speed Internet access from space. In May, the company will launch test versions of the Kuiper communications satellites in an attempt to take on SpaceX's own venture, Starlink, and tap into a market of perhaps hundreds of millions of prospective internet users.

Other companies are hoping to do the same, and a few are already doing so, but Starlink and Amazon are the major players. "It is really a head-to-head rivalry," says Tim Farrar, a satellite expert from the firm TMF Associates in the U.S. 

The rocket that will launch Amazon's first two Kuiper satellites—the United Launch Alliance's new Vulcan Centaur rocket—has been assembled at Cape Canaveral in Florida. Its inaugural launch is set to fly two prototype Kuiper satellites, called KuiperSat-1 and KuiperSat-2, as early as May 4. Ultimately, Amazon plans to launch a total of 3,236 full Kuiper satellites by 2029. The first of that fleet could launch in early 2024.

"They have ambitions to be disruptive across the technology sector," says Farrar. "It's hardly surprising that they've jumped in here."

From MIT Technology Review

Wharton AI Powered Risk Radar and more

Wharton AI Powered Risk Radar"  From Bard

Wharton researchers have developed a new AI-powered tool to help businesses identify and mitigate risks. The tool, called "Risk Radar," uses machine learning to analyze data from a variety of sources, including financial statements, news reports, and social media. Risk Radar then identifies potential risks that businesses may face, such as fraud, financial instability, and reputational damage.

Risk Radar is still in development, but it has already been used by several businesses to identify and mitigate risks. For example, one business used Risk Radar to identify a potential fraud scheme that was targeting its customers. The business was able to stop the scheme before it caused any damage.

Wharton researchers are continuing to develop Risk Radar and are working to make it available to more businesses. Risk Radar is a powerful tool that can help businesses identify and mitigate risks, and it has the potential to save businesses millions of dollars.

Wharton researchers have developed an AI-powered tool to help businesses improve their marketing campaigns. The tool, called "Marketing AI," uses machine learning to analyze customer data, such as customer interactions, website visits, and social media posts. Marketing AI then identifies potential areas where businesses can improve their marketing campaigns, such as areas where customers are likely to respond to certain types of marketing messages.

Wharton researchers have developed an AI-powered tool to help businesses improve their financial analysis. The tool, called "Financial Analysis AI," uses machine learning to analyze financial data, such as financial statements, stock prices, and economic data. Financial Analysis AI then identifies potential areas where businesses can improve their financial analysis, such as areas where businesses can identify potential risks or opportunities.

Wharton researchers have developed an AI-powered tool to help businesses improve their human resources management. The tool, called "Human Resources AI," uses machine learning to analyze employee data, such as employee performance reviews, employee surveys, and employee turnover data. Human Resources AI then identifies potential areas where businesses can improve their human resources management, such as areas where businesses can identify potential employee satisfaction or retention issues.

These are just a few examples of how Wharton is using AI to improve the way businesses operate. As AI continues to develop, Wharton will continue to find new ways to use it to help businesses succeed.


Monday, March 27, 2023

Crewless Container Ships

Saw several example that indicated they were in test.  Minimal crew still might add safety at low cost for a large ship? 

Crewless Container Ships Appear on the Horizon

By BBC News,  March 27, 2023

Eventually will navigate aided by sensors, including radar and cameras, which will feed data to an artificial intelligence, which will detect and classify waterborne obstacles.

The Norwegian container ship Yara Birkeland eventually will become a crewless vessel that navigates via radar and cameras, feeding sensor data to artificial intelligence that will detect and identify obstacles in the water.  Its captain will remotely monitor the ship at a land-based facility more than 80 kilometers (50 miles) off, ready to intervene if necessary.

Norway-based Kongsberg is providing technology for the Yara Birkeland and two other autonomous barges in the Oslo Fjord.  The University of South-Eastern Norway's Marius Tannum said unmanned vessels should be backed up by human monitors, with crews remaining on board for a transitional period.

Tannum expects autonomous shipping's momentum to overtake that of autonomous vehicles, because crewless vessels "with a fixed route and a remote operation center [ROC] will be operating with less risk than unmanned autonomous trucks driving in regular traffic."

From BBC News

View Full Article    

Privacy Flaw in ChatGPT

Schneier reports on this,    as usual his commenters provide thoughtful additional insight.  Worh looking at.

ChatGPT Privacy Flaw

OpenAI has disabled ChatGPT’s privacy history, almost certainly because they had a security flaw where users were seeing each others’ histories.   

Tags: ChatGPT, cybersecurity, privacy

Posted on March 22, 2023 at 7:14 AM • 16 Comments

AI for Materials Discovery

New, lucrative  and very interesting space

Artificial Intelligence for Materials Discovery

By Don Monroe

Communications of the ACM, April 2023, Vol. 66 No. 4, Pages 9-11   10.1145/3583080

3D chemical compounds floating in space, illustration  ... 

The software-driven successes of deep learning have been profound, but the real world is made of materials. Researchers are turning to artificial intelligence (AI) to help find new materials to provide better electronics and transportation, and the energy to run them.

Despite its undeniable power, however, "Machine learning, especially the deep learning revolution, relies heavily on large amounts of data," said Carla Gomes, a computer scientist at Cornell University. "This is not how science works. [Scientists] don't just memorize things."

"Machine learning as we know it is not enough for scientific discovery," she said. "We still have a long way to go."

Nevertheless, researchers are off to a promising start in addressing materials science.

Combinatorial Explosion

One of the challenges in materials discovery is the astronomical number of compositions that might have interesting properties. "High-entropy alloys" (HEA), for example, combine four or more metals. "If you consider all the elements in the periodic table and you will find that you have many combinations, then infinite combinations of the different elements, so that makes prediction very difficult," explained Ziyuan Rao, a postdoc at the Max Planck Institute for Iron Research in DĂĽsseldorf, Germany.

Nonetheless, Rao and his colleagues created a multistage analysis to search for alloys with low thermal expansion, which are important for cryogenic storage of liquified natural gas and for other purposes. The analysis draws on extensive materials datasets, but the available compositions are a tiny, sparse subset of the universe of perhaps 1050 possibilities.

After training a machine-learning model with this data, the researchers used it to select promising candidates, often completely novel. They then used computationally intensive density-functional theory (DFT) calculations to get more precise estimates of each compound's properties. DFT is a widely used shortcut around full quantum mechanical theory. In fact, researchers at DeepMind recently used deep learning to let DFT determine how electron charge is distributed between competing atoms, a longstanding challenge.

A key feature of the HEA search is active learning, which suggests new compositions to examine that will be most informative. "it's a little different from traditional machine learning," Rao said, which typically aims to increase the accuracy of the model. "We also want to use this model to predict new materials with very good properties."

Indeed, Rao and his colleagues further refined their search by experimentally making and measuring some of the best candidates. "You need real-world data," he said, because "Simulated data sometimes is inaccurate." The experimental results are folded back into the modeling, and the loop is repeated six times. The study successfully identified two new alloy compositions with a tiny thermal expansion coefficient, less than two parts per million per degree.  ... ' 

More on Plugins and ChatGPT

 Intro below,   saw the intro from wolfram. Am inclined to think this is a big deal,   Like Apps within searches.    Will be exploring this in the coming weeks.

What are ChatGPT plugins? Here’s everything you need to know  from PocketNow

By SANUJ BHATIA, OpenAI has been making headlines since the end of last year, primarily due to the immense popularity of its flagship service, ChatGPT. This AI-powered chatbot has taken the world by storm and has proven to be a useful tool for many people. The company has been on a run of announcements, recently unveiling its GPT-4 model, and has now introduced ChatGPT plugins. In this article, we’ll explain what ChatGPT plugins are, what they can do for you, and what plugins are available right now.

What are ChatGPT Plugins?

One of the limitations of ChatGPT is that it can reply to a user's query based only on the training data it has, which is limited to 2021. This means that ChatGPT is unaware of the latest events or even those that occurred in the last year. Plugins will essentially allow ChatGPT to access the world of the internet and retrieve information from it.

Think of plugins as apps for ChatGPT

Plugins will allow the service to interact with live data from the web and specific websites. Think of plugins as apps for ChatGPT. Using these plugins, the AI chatbot will be able to perform a number of tasks that it has not been able to do until now. OpenAI says that plugins are like ChatGPT's "eyes and ears" and have the potential to turn the chatbot into a versatile interface for a variety of services and websites.

Using the plugins, ChatGPT will be able call APIs of compatible third-party services to perform actions. In a post, the company wrote "For instance, if a user asks, "Where should I stay in Paris for a couple nights?", the model may choose to call a hotel reservation plugin API, receive the API response, and generate a user-facing answer combining the API data and its natural language capabilities."

Plugins for ChatGPT use

Good piece on current Open ChatGPT Plugins, including the ability to join the waitlist for their use. 

ChatGPT plugins  https://openai.com/blog/chatgpt-plugins 

We’ve implemented initial support for plugins in ChatGPT. Plugins are tools designed specifically for language models with safety as a core principle, and help ChatGPT access up-to-date information, run computations, or use third-party services. ... ' 

Where Is the Research on Cryptographic Transition and Agility?

VIEWPOINT

Where Is the Research on Cryptographic Transition and Agility?

By David Ott, Kenny Paterson, Dennis Moreau

Communications of the ACM, April 2023, Vol. 66 No. 4, Pages 29-32   10.1145/3567825

Shield-like structure with image of a key at its center, illustration

As quantum computing technology continues to advance, the industry is faced with the challenge of migrating to new, quantum-safe public key cryptography standards. Based on algorithms known collectively as post-quantum cryptography, these standards are actively under development by the U.S. National Institute of Standards and Technology (NIST) in collaboration with the broader cryptography research community.4

This is not the first time a transition to new cryptography has been needed. Algorithmic vulnerabilities, more powerful hardware platforms, more efficient algorithms, amongst other reasons, have motivated prior transitions in cryptographic hash algorithms (MD5 to SHA1, SHA1 to SHA2), symmetric key algorithms (RC4 to DES to AES), and public key algorithms (RSA-1024 to -2048, RSA to ECC). In fact, cryptographic transitions include not only algorithms, but protocols applying those algorithms (SSL/TLS, KMIP, IKE) and the actual implementations (OpenSSL, Bouncy Castle) that are the workhorse for deployed solutions.  ... '

Configurable Robots Can Be Mixed, Matched for Tasks in Space

Configurable Robots Can Be Mixed, Matched for Tasks in Space

ACM TECHNEWS

By Interesting Engineering

March 22, 2023

The Walking Oligomeric Robotic Mobility System could revolutionize how we approach space robotics, according to MIT.

Credit: MIT

Massachusetts Institute of Technology (MIT) researchers engineered the Walking Oligomeric Robotic Mobility System (WORMS) to revolutionize space robotics.

MIT's Brooke Bensche said, "Our idea was that, with just a few parts, combined in different ways, you could mix and match and get all these different robots."

The university's Michael Brown said the system's appendages were inspired by the humble worm because each can make "similar movements as an arm, or a leg, or a backbone, or a tail."

Each configurable articulated robot carries its own motors, sensors, computer, and battery; they can be assembled individually to perform small tasks or together for more complex functions.

From Interesting Engineering

View Full Article  

Towards Talking Concrete

Novel infrastructure idea.

'Talking' Concrete Could Help Prevent Traffic Jams, Cut Carbon Emissions

By Purdue University News

March 16, 2023

From beneath a concrete pour, this black circular sensor transmits data about the concrete's strength levels through a cord plugged into an above-ground handheld device called a data logger.

Credit: Rebecca McElhoe/Purdue University

U.S. interstates are preparing to test sensors developed by Purdue University researchers that could help prevent congestion and lower carbon emissions.

The sensors enable concrete pavement to relay data about its strength and repair requirements to engineers, forgoing the need to test samples; this should reduce construction time and repair frequency while cutting emissions from vehicles waiting to bypass construction sites.

The devices communicate to engineers through a smartphone application when the pavement is sufficiently strong to tolerate heavy traffic.

More than half of U.S. states with concrete interstate pavement will deploy the sensors as part of a Federal Highway Administration study, with Indiana and Texas already testing them in highway paving projects.

The WaveLogix company will manufacture the technology for commercialization as the REBEL Concrete Strength Sensing System.

From Purdue University News

View Full Article


Sunday, March 26, 2023

Machine Learning Street talk: Wolfram Announcement

I mentioned the below announcement of being able to plug in WolframAlpha capabilities in ChatGPT.  I found the the talk here to be insightful about how language models could interact with computational models.  And how can this be used yet further improve AI.  Technical, but could be pointing us to yet more wonderful things.

Machine Learning Street Talk    https://youtu.be/z5WZhCBRDpU

132,828 views  Mar 23, 2023  #110  Episode #110

HUGE ANNOUNCEMENT, CHATGPT+WOLFRAM! You saw it HERE first!

ChatGPT + Wolfram: The Future of AI is Here!

Pod version: https://podcasters.spotify.com/pod/sh...

Support us! https://www.patreon.com/mlst 

MLST Discord: https://discord.gg/aNPkGUQtc5

Stephen's announcement post: https://writings.stephenwolfram.com/2... 

OpenAI's announcement post: https://openai.com/blog/chatgpt-plugins 

In an era of technology and innovation, few individuals have left as indelible a mark on the fabric of modern science as our esteemed guest, Dr. Steven Wolfram. 

Dr. Wolfram is a renowned polymath who has made significant contributions to the fields of physics, computer science, and mathematics. A prodigious young man too, Wolfram earned a Ph.D. in theoretical physics from the California Institute of Technology by the age of 20. He became the youngest recipient of the prestigious MacArthur Fellowship at the age of 21.

Wolfram's groundbreaking computational tool, Mathematica, was launched in 1988 and has become a cornerstone for researchers and innovators worldwide. In 2002, he published "A New Kind of Science," a paradigm-shifting work that explores the foundations of science through the lens of computational systems.

In 2009, Wolfram created Wolfram Alpha, a computational knowledge engine utilized by millions of users worldwide. His current focus is on the Wolfram Language, a powerful programming language designed to democratize access to cutting-edge technology.

Wolfram's numerous accolades include honorary doctorates and fellowships from prestigious institutions. As an influential thinker, Dr. Wolfram has dedicated his life to unraveling the mysteries of the universe and making computation accessible to all.

First of all... we have an announcement to make, you heard it FIRST here on MLST! ....

[00:00] Intro

[02:57] Big announcement! Wolfram + ChatGPT!

[05:33] What does it mean to understand?

[13:48] Feeding information back into the model

[20:09] Semantics and cognitive categories

[23:50] Navigating the ruliad

[31:39] Computational irreducibility

[38:43] Conceivability and interestingness

[43:43] Human intelligible sciences

Gates Agrees about Importance of Current AI

 More agreement about AI advances.

Bill Gates: AI is most important tech advance in decades

By Tom Gerken   BBC Technology reporter

Microsoft co-founder Bill Gates says the development of artificial intelligence (AI) is the most important technological advance in decades.

In a blog post on Tuesday, he called it as fundamental as the creation of the microprocessor, the personal computer, the Internet, and the mobile phone.

"It will change the way people work, learn, travel, get health care, and communicate with each other," he said.

He was writing about the technology used by tools such as chatbot ChatGPT.

Developed by OpenAI, ChatGPT is an AI chatbot which is programmed to answer questions online using natural, human-like language.

The team behind it in January 2023 received a multibillion dollar investment from Microsoft - where Mr Gates still serves as an advisor.

But it is not the only AI-powered chatbot available, with Google recently introducing rival Bard. ... ' 

Metal Detecting from the Air

Recall a DOD project looking a a related problem.   Hope this does not become an issue in greater Europe.

Metal-Detecting Drone Could Autonomously Find Land Mines A drone with 5 degrees of freedom can safely detect buried objects from the air.  By Evan Ackerman 

Several stitched-together photographs of gray three-propellered drones with metal detectors hovering just above grass

This composite photo shows how a tricopter drone with a lidar and metal detector can fly around an obstacle close to the ground.

Metal detecting can be a fun hobby, or it can be a task to be completed in deadly earnest—if the buried treasure you’re searching for includes land mines and explosive remnants of war. This is an enormous, dangerous problem: Something like 12,000 square kilometers worldwide are essentially useless and uninhabitable because of the threat of buried explosives, and thousands and thousands of people are injured or killed every year.

While there are many different ways of detecting mines and explosives, none of them are particularly quick or easy. For obvious reasons, sending a human out into a minefield with a metal detector is not the safest way of doing things. So, instead, people send anything else that they possibly can, from machines that can smash through minefields with brute force to well-trained rats that take a more passive approach by sniffing out explosive chemicals.

Because the majority of mines are triggered by pressure or direct proximity, it may seem that a drone would be the ideal way to detect them nonexplosively. However, unless you’re only detecting over a perfectly flat surface (and perhaps not even then) your detector won’t be positioned ideally most of the time, and you might miss something, which is not a viable option for mine detection.

But now a novel combination of a metal detector and a drone with 5 degrees of freedom is under development at the Autonomous Systems Lab at ETH Zurich. It may provide a viable solution to remote land-mine detection, by using careful sensing and localization along with some twisting motors to keep the detector reliably close to the ground.  ...' 

A Legal Challenge to Algorithmic Recommendations

Regulation required.

A Legal Challenge to Algorithmic Recommendations

By Pamela Samuelson

Communications of the ACM, March 2023, Vol. 66 No. 3, Pages 32-34

Credit: Andrij Borys Associates

A young American student, Nohemi Gonzalez, was one of 149 people murdered in Paris in 2015 by ISIS terrorists. Her family blames Google for her death, claiming that YouTube's algorithms provided material support to the terrorist organization by recommending violent and radicalizing ISIS videos to its users based on their previous viewing histories. (The Gonzalez complaint levies the same charges against Twitter and Facebook, but to keep things simple, this column refers only to Google.)

Gonzalez' family sued Google for damages for this wrongful death. Both a trial and an appellate court agreed with Google that it could not be held liable for this tragic death under a federal immunity shield widely known as § 230 of the Communications Decency Act (CDA). However, the U.S. Supreme Court has decided to hear Gonzalez' appeal and consider whether YouTube's algorithmic recommendations are beyond the shelter of § 230.... '


AI Godfather says New Tech Could be Dangerous

 Well known AI Scientist Geoffrey Hinton,  often mentioned here,  who we followed closely n the 80s, says AI is real, and conceivably threatening humanity.    Equivalent to Invention of Wheel or electricity.  Scary statements from him. 

Artificial intelligence 'godfather' on AI possibly wiping out humanity: ‘It's not inconceivable’

Artificial Intelligence pioneer Geoffrey Hinton said the development of artificial intelligence is happening rapidly

Andrea VacchianoBy Andrea Vacchiano | Fox News

Geoffrey Hinton, a computer scientist who has been called "the godfather of artificial intelligence", says it is "not inconceivable" that AI may develop to the point where it poses a threat to humanity.

The computer scientist sat down with CBS News this week about his predictions for the advancement of AI. He compared the invention of AI to electricity or the wheel.

Hinton, who works at Google and the University of Toronto, said that the development of general purpose AI is progressing sooner than people may imagine. General purpose AI is artificial intelligence with several intended and unintended purposes, including speech recognition, answering questions and translation.

"Until quite recently, I thought it was going to be like 20 to 50 years before we have general purpose AI. And now I think it may be 20 years or less," Hinton predicted. Asked specifically the chances of AI "wiping out humanity," Hinton said, "I think it's not inconceivable. That's all I'll say." 

Geoffrey Hinton, chief scientific adviser at the Vector Institute, speaks during The International Economic Forum of the Americas (IEFA) Toronto Global Forum in Toronto, Ontario, Canada, on Thursday, Sept. 5, 2019. 

Geoffrey Hinton, chief scientific adviser at the Vector Institute, speaks during The International Economic Forum of the Americas (IEFA) Toronto Global Forum in Toronto .. 

Artificial general intelligence refers to the potential ability for an intelligence agent to learn any mental task that a human can do. It has not been developed yet, and computer scientists are still figuring out if it is possible.  Hinton said it was plausible for computers to eventually gain the ability to create ideas to improve themselves. 

"That's an issue, right. We have to think hard about how you control that," Hinton said.   ... ' 

Saturday, March 25, 2023

Large Language Models: A Cognitive and Neuroscience Perspective

Irving does an excellent review and links to much work about LLMs, Large Language Models, and oter topics that are now much in the news. Below an intro. I plan to read all the articles pointed to at the link.  A considerable weakness in the current directions?   Implications to all this,  will provide.

A collection of observations, news and resources on the changing nature of innovation, technology, leadership, and other subjects.  By Irving Wladawsky-Berger  March 23, 2023

Large Language Models: A Cognitive and Neuroscience Perspective

Over the past few decades, powerful AI systems have matched or surpassed human levels of performance in a number of tasks such as image and speech recognition, skin cancer classification, breast cancer detection, and highly complex games like Go. These AI breakthroughs have been based on increasingly powerful and inexpensive computing technologies, innovative deep learning (DL) algorithms, and huge amounts of data on almost any subject. More recently, the advent of large language models (LLMs) are taking AI to the next level. And, for many technologists like me, LLMs and their associated chatbots have introduced us to the fascinating world of human language and cognition.

I recently learned the difference between form, communicative intent, meaning, and understanding from “Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data,” a 2020 paper by linguistic professors Emiliy Bender and Alexander Koller. These linguistic concepts helped me understand the authors’ argument that “in contrast to some current hype, meaning cannot be learned from form alone. This means that even large language models such as BERT do not learn meaning; they learn some reflection of meaning into the linguistic form which is very useful in applications.”

A few weeks ago, I came across another interesting paper, “Dissociating Language and Thought in Large Language Models: a Cognitive Perspective,” published in January, 2023 by principal authors linguist Kyle Mahowald and cognitive neuroscientist Anna Ivanova and four additional co-authors. The paper nicely explains how the study of human language, cognition and neuroscience sheds light on the potential capabilities of LLMs and chatbots. Let me briefly discuss what I learned.

“Today’s large language models (LLMs) routinely generate coherent, grammatical and seemingly meaningful paragraphs of text,” said the paper’s abstract. “This achievement has led to speculation that these networks are — or will soon become — thinking machines, capable of performing tasks that require abstract knowledge and reasoning. “Here, we review the capabilities of LLMs by considering their performance on two different aspects of language use: formal linguistic competence, which includes knowledge of rules and patterns of a given language, and functional linguistic competence, a host of cognitive abilities required for language understanding and use in the real world.

The authors point out that there’s a tight relationship between language and thought in humans. When we hear or read a sentence, we typically assume that it was produced by a rational person based on their real world knowledge, critical thinking, and reasoning abilities. We generally view other people’s statements not just as a reflection of their linguistic skills, but as a window into their mind. .... '


Capturing What is Said

The value of good speech to text. 

Capturing What is Said,  By Esther Shein

Commissioned by CACM Staff, March 23, 2023

A very basic flow chart for the conversion of speech to text.

New AI-enabled capabilities for speech-to-text systems include taking actions based on a transcript, prompting someone to ask a follow-up question, and summarizing a conversation at the end of a call, said Christine McAllister at Forrester Research.

ChatGPT and generative artificial intelligence (AI) may be having a moment, but don't underestimate the value of speech-to-text transcription, sometimes referred to as automatic speech recognition (ASR) software, which continues to improve.

ASR technology converts human speech into text using machine learning and AI. There are two types: synchronous transcription, which is typically used in chatbots, and asynchronous, where transcription occurs after the fact to capture customer/agent conversations, notes Cobus Greyling, chief evangelist at HumanFirst, which makes a productivity suite for natural language data.

ASR made some waves in recent months with the announcement of Whisper from OpenAI, the organization that created ChatGPT. Whisper was trained on 680,000 hours of multilingual and supervised data collected from the Web. OpenAI claims that large and diverse dataset has improved the accuracy of the text it produces; the company says Whisper also can transcribe text from speech in multiple languages.

"What that means is that it's extremely accurate—right off the top—without much tuning or training,'' says Christina McAllister, a senior analyst at research and advisory company Forrester Research. "The large language model aspect, which is based on huge amounts of data, is what's new and is the most innovative aspect of the ASR market today,'' she says.

Because of its ability to transcribe meetings and interviews more efficiently and accurately, one of the broadest enterprise use cases for speech-to-text is in customer call centers. The next phase in the development of ASR is to use artificial intelligence to analyze call center conversations for customer sentiment and to validate compliance in regulated industries, according to Annette Jump, a vice president analyst at Gartner.

The benefits of ASR in the call center context are its ability to identify customer problems early and to improve customer satisfaction by resolving issues sooner, says Jump.

Other use cases include generating closed captions for movies, television, video games, and other forms of media. ASR is widely used in healthcare by physicians to convert dictated clinical notes into electronic medical records.

Speech vendors typically leverage a third-party ASR engine so they don't have to build their own, McAllister says. That frees them up so they can "do all the rest of their magic from the transcript point forward,'' she says.

Some of the new AI capabilities for speech-to-text systems include taking actions based on a transcript, prompting someone when it's appropriate to ask a follow-up question, and summarizing a conversation at the end of a call, McAllister says.

One frequently used AI-powered speech-to-text transcription service is Otter.ai, which has added capabilities aimed at improving meetings, including integration with collaboration tools such as Zoom and Microsoft Outlook.  ... ' 


Gordon Moore, Intel co-founder and creator of Moore's Law, dies aged 94

Gordon Moore, Intel co-founder and creator of Moore's Law, dies aged 94

Published, 14 hours ago  in the BBC

Silicon Valley pioneer and philanthropist Gordon Moore has died aged 94 in Hawaii.

Mr Moore started working on semiconductors in the 1950s and co-founded the Intel Corporation.

He famously predicted that computer processing powers would double every year - later revised to every two - an insight known as Moore's Law.

That "law" became the bedrock for the computer processor industry and influenced the PC revolution.

Two decades before the computer revolution began, Moore wrote in a paper that integrated circuits would lead "to such wonders as home computers - or at least terminals connected to a central computer - automatic controls for automobiles, and personal portable communications equipment".

He observed, in the 1965 article, that thanks to technological improvements the number of transistors on microchips had roughly doubled every year since integrated circuits were invented a few years earlier.

His prediction that this would continue became known as Moore's Law, and it helped push chipmakers to target their research to make this come true.

After Moore's article was published, memory chips became more efficient and less expensive at an exponential rate. ...   '

Would a TikTok Ban Cripple Influencer Marketing?

Been following this recently on Youtube, novelty key.  Much more by marketing experts at the link.

Would a TikTok Ban Cripple Influencer Marketing?

Mar 21, 2023,  by Tom Ryan

The Biden administration last week threatened to ban TikTok in the U.S. if the app’s Chinese owners refuse to sell their stakes, raising questions about if competitors Instagram, Snap and YouTube can fill the void for the influencer community.

For the third year in a row, TikTok held the record as the most downloaded app with nearly 40 percent of TikTok’s advertising audience aged between 18 and 24.

Influencer Marketing Hub’s “The State of Influencer Marketing 2023” report found that TikTok is now the most popular influencer marketing channel (utilized by 56 percent of brands using influencer marketing), surpassing Instagram (51 percent) for the first time. Facebook (42 percent) and YouTube (38 percent) follow behind.

Thomas Walters, Europe CEO and co-founder of creator agency Billion Dollar Boy, told Campaign that factors such as TikTok’s personalized algorithm, lack of emphasis on follower count and its shift towards “spontaneous, raw, unfiltered content” are all helping drive the platform’s popularity.

In a recent blog entry, Kolsquare, the influencer marketing platform, said that while facing challenges with conversion, TikTok is “the clear leader when it comes to driving trends and engagement amongst the youngest users of social media” and has an edge in driving awareness over other platforms.

Both the Trump and Biden administrations have said that the app poses a national security threat amid concerns China could tap user data to spread misinformation. On March 23, testimony by TikTok CEO Shou Zi Chew before the House Energy and Commerce committee may raise the rhetoric.

Some content creators are seeking to diversify to Instagram Reels or YouTube Shorts, but others doubt the TikTok ban will take effect, according to The Wall Street Journal.

In terms of potential purchasers for the app, a New York Times article speculated that many buyers could not afford TikTok (valuation at $50 billion or more) or would not want to deal with the antitrust scrutiny of an acquisition.

Many brands, according to Advertising Age, continue to put marketing dollars behind TikTok, given the likely delays or challenges that would come with enacting an outright ban. Becca Millstein, CEO of tinned fish brand Fishwife, told Adage, “We believe that TikTok can be a powerful source of organic discovery and hope that we have the opportunity to utilize it as such.”  ... '


Yes, ChatGPT Is Coming for Your Office Job

Have now seen a number of posts brainstorming this,  and considering what I have seen, there re many reasonable ideas, and in particularly easy to test with even current systems.  Only overstepping regulation could get in the way,  and even that could be reasonably addressed.  Likely GPT based systems will be seen everywhere.  Agree the disruption may not be the what we expect.

Yes, ChatGPT Is Coming for Your Office Job,     By Wired,  March 23, 2023

With companies like Microsoft, Slack, and Salesforce adding ChatGPT or similar AI tools to their products, we are likely to see the impact on office life soon enough.

Anyone who has spent a few minutes playing with ChatGPT will understand the worries and hopes such technology generates when it comes to white-collar work. The chatbot is able to answer all manner of queries—from coding problems to legal conundrums to historical questions—with remarkable eloquence.

Assuming companies can overcome the problematic way these models tend to "hallucinate" incorrect information, it isn't hard to imagine they might step in for customer support agents, legal clerks, or history tutors. Such expectations are fueled by studies and media reports claiming that ChatGPT can get a passing grade on some legal, medical, and business exams. With companies like Microsoft, Slack, and Salesforce adding ChatGPT or similar AI tools to their products, we are likely to see the impact on office life soon enough.

A couple of research papers posted online this week suggest that ChatGPT and similar chatbots may be very disruptive—but not necessarily in the ways you expect.

From Wired

View Full Article  


Google Glass is Over

 Too bad, tried it and it was nicely done. Is the category over for now?   Has to be a place for this. 

Google Glass is finally shattered   in Popsci

10 years after its debut, Google finally shutters the headset for good.  By Andrew Paul

What happened to Google Glass?

RIP Google Glass (2013-2023). LEA SUZUKI/THE SAN FRANCISCO CHRONICLE VIA GETTY IMAGES    

Back in 2013, Google attempted to get way ahead of the augmented reality game with its Google Glass headset. Although first billed as a tech game changer, the $1,500 price tag and privacy concerns made it a wholesale commercial flop. But despite some belated success among medical professionals and first responders via Google’s 2017 Glass Enterprise revamp, the much-memed product never really broke through to the masses. Yesterday, Google officially announced the demise of its Glass product line.

According to a company statement, headsets are no longer available for purchase, while support for Glass Enterprise Edition will continue through mid-September of this year. “Thank you for over a decade of innovation and partnership,” writes Google, in the brief end to one of the more infamous modern tech rollouts.

[Related: Doctors are wearing the new Google Glass while seeing patients.]

Initially resembling frameless eyewear, Google’s headset included a small, rectangular, transparent glass (hence the name) above the wearer’s right eye. A miniature onboard computer system beamed bits of information through the prism. Users could then utilize features like map directions, photo and video capabilities, and weather forecasts in front of them while maintaining a clear vision of their surrounding, real-world environment. Future iterations resembled protective eyewear designs, and were often utilized in industries such as factory manufacturing.

The announcement likely comes as no surprise to most—alongside a “Wait, Google still made those?” from many others—as the Big Tech giant’s last edition of Glass Enterprise came out almost four years’ ago in 2019 alongside a $999 price tag. Since then, Google’s chief rivals at Meta and Apple have poured massive amounts of cash into their own respective AR projects. In 2021, Meta collaborated with Ray-Ban to release camera-embedded sunglasses, albeit with no augmented display features, and (until recently) was going all in on pushing a “metaverse” experience via its Meta Quest headset line. Meanwhile, Apple is widely reported to be on the cusp of rolling out its own wearable AR/VR product line....' 

Security in Cyberspace

Considerable piece in Fraunhofer mag,  intro below

More security in cyberspace

Web special Fraunhofer magazine 2.2022

The invasion of Ukraine shows that fighting is no longer just on the battlefield, but also in virtual space - with highly professional hacker attacks and targeted disinformation. How can Germany become more defensive?

Months before Putin gave his troops the marching orders, the war on the Internet began. Hackers have been preparing the Russian invasion since at least December of last year. This is the conclusion of Prof. Haya Shulman, who studied the cyber attacks on Ukrainian infrastructure. Shulman heads the "Cybersecurity Analytics and Defences" department at the Fraunhofer Institute for Secure Information Technology SIT in Darmstadt, coordinates the "Analytics Based Cybersecurity" research area at the National Research Center for Applied Cybersecurity ATHENE and holds a chair at the Goethe University in Frankfurt am Main.

One of their findings: The malware that caused the systems of the communications satellite KA-SAT to fail on the day Russia invaded Ukraine was smuggled in months ago. KA-SAT provides broadband internet to customers across Europe and is used by the Ukrainian Army for emergency communications. »The goal of the hackers was to stop the communication – and they succeeded.

It took a month for the damage to be repaired, at least for the most part,” says Shulman. The Russian attack on KA-SAT was also not without consequences in Germany and all of Central Europe: 5,800 wind turbines could no longer be maintained and controlled remotely. Systems in remote locations that are connected to the Internet via a satellite connection were affected. They continued to supply electricity. However, technical problems could only be identified and rectified on site.  ...... ' 

Researchers Develop Soft Robot That Shifts from Land to Sea with Ease

Mobile robotics in multiple domains.

ACM TECHNEWS

Researchers Develop Soft Robot That Shifts from Land to Sea with Ease

By Carnegie Mellon University, March 21, 2023

Actuators allow the reconfigurable robot to curl its body to swiftly roll away.

Credit: Morphing Matter Lab

Soft robots developed by researchers at Carnegie Mellon University can transition from walking to swimming or crawling to rolling, shifts found in most animals.

The researchers created a bistable actuator using three-dimensionally printed soft rubber with alloy springs that contract in response to electrical currents, allowing the actuator to bend. The robot remains in the new shape until it reverts to its previous configuration in response to another electrical charge. Only a hundred milliseconds of electrical charge is needed to change shape.

The researchers created robots that can walk and swim, crawl and jump, and crawl and roll.

"Our bistable actuator is simple, stable and durable, and lays the foundation for future work on dynamic, reconfigurable soft robotics," says Dinesh K. Patel, a post-doctoral fellow in CMU's Morphing Matter Lab.

From Carnegie Mellon University

View Full Article  

Quantum Computers May Finally Have Practical Use

 A means to generate truly random numbers for cryptology

ACM TECHNEWS

Quantum Computers May Finally Have Practical Use

By New Scientist, March 24, 2023

Google's Sycamore quantum computer. 

Quantum computers such as Google’s Sycamore could be put to use creating numbers that are guaranteed to be truly random.

University of Texas at Austin (UT Austin) researchers have developed a method for certifying that quantum computers generate truly random numbers without having to inspect the process.

This involves asking a quantum computer to complete a test in which a series of pseudorandom operations are run on its qubits and measuring the outputs, which act as truly random numbers.

If the resulting outputs cannot be simulated on a classical computer, they are confirmed to be the result of quantum processes, truly random, and suitable for cybersecurity applications.

Said UT Austin's Scott Aaronson, "The huge advantage with this proposal is that you can actually do it with devices that currently exist."

From New Scientist

View Full Article - May Require Paid Subscription   

Friday, March 24, 2023

Plugins to be Added to ChatGPT, to Create a Platform

Announcement relates to previous message on Wolfram Plugins, but promises more yet.

OpenAI turns ChatGPT into a platform overnight with addition of plugins  in Venturebeat

Sharon Goldman@sharongoldman,     Michael Nuñez@MichaelFNunez    March 23, 2023 

OpenAI booth at NeurIPS 2019 in Vancouver, Canada

OpenAI today announced its support of new third-party plugins for ChatGPT, and it already has Twitter buzzing about the company’s potential platform play.

In a blog post, the company stated that the plugins are “tools designed specifically for language models with safety as a core principle, and help ChatGPT access up-to-date information, run computations, or use third-party services.”  ... 

A sign of OpenAI’s accelerating dominance

The announcement was quickly received by the public as a signal of OpenAI‘s ambitions to further its dominance by turning ChatGPT into a developer platform.

“OpenAI is seeing ChatGPT as a platform play,” tweeted Marco Mascorro, cofounder of Fellow AI.

Join us in San Francisco on July 11-12, where top executives will share how they have integrated and optimized AI investments for success and avoided common pitfalls.

And @gregmushen tweeted: “I think the introduction of plugins to ChatGPT is a threat to the App Store. It creates a new platform with new monetization methods.”

In sharing the announcement, OpenAI CEO Sam Altman tweeted: “We are starting our rollout of ChatGPT plugins. you can install plugins to help with a wide variety of tasks. we are excited to see what developers create!”    ... ' 

What Does it Mean to be Smart in an Age of AI

Excerpt from McKinsey  Complete video at the link.

Author Talks: In the ‘age of AI,’ what does it mean to be smart?

March 16, 2023 | Interview

As artificial intelligence gets better at predicting human behavior, a business psychologist encourages people to strengthen the uniquely human skills that machine learning has yet to tap.

In this edition of Author Talks, McKinsey Global Publishing’s Raju Narisetti chats with Tomas Chamorro-Premuzic about his new book, I, Human: AI, Automation, and the Quest to Reclaim What Makes Us Unique (Harvard Business Review Press, February 2023). Chamorro-Premuzic explains why some AI algorithms model humanity as a simple species, how attention has become commoditized, and why the right questions are now more valuable than the right answers. An edited version of the conversation follows.

Why did you write this, your 12th book, now?

An unanticipated problem was encountered, check back soon and try again

Video

I’m a professor of business psychology at Columbia University and UCL [University College London] and the chief innovation officer at ManpowerGroup. I, Human: AI, Automation, and the Quest to Reclaim What Makes Us Unique is a book about the behavioral consequences or impact of artificial intelligence, including the dark side of human behavior and what we should do to upgrade ourselves as a species.

The book is written at a time that, in my view, could only be described as the AI age. Humans have always relied on technological inventiveness and innovation to shape their cultural and social evolution, and I think there can be very little doubt that the definitive technology of today is artificial intelligence, or AI.

Now, even the wider public is talking about things like ChatGPT and other conversational interfaces, and the tech giants are described mostly as data companies and as algorithmic prediction businesses.

The book was very much written in the midst of the AI age, or under the influence of AI, because I wrote the bulk of this at the height of the pandemic when we had very little physical interaction or contact with other people outside of our nuclear families. This means I was heavily influenced by hyperconnectedness and the datafication of me. Everything I did was being datafied and subjected to the predictive powers of AI during 2020 and 2021.

I can’t say that there won’t be a better era to read the book, but it certainly wouldn’t have had the same connotation and impact if we had published it five or ten years ago.

Haven’t humans always blamed technology for every problem they face?

There is a common tendency for people to overreact to things that are novel, whether in a good way or in a bad way, and technologies are a very good example of this.

Perhaps the best example is how, when the written newspaper first scaled up and productized, people feared that humans would never meet in person ever again because there would be no information or even gossip to exchange if all the news was in written form. Also, from the 1950s onward, people showed concern that television would lead to less intellectual activities, but I don’t think they were wrong because reading habits went down since mass TV was introduced.

What I tried to do with this book is not be at one extreme or the other. What’s important to me is to not miss the opportunity to highlight the behavioral impact and consequences that we have already seen artificial intelligence have on us. This is not a book about AI, but about humans in the AI age.  ... '