AI Watch
Source Link
Excerpt:
Bottom line: As top labs race to build an AI master race, many turn a blind eye to dangerous behaviors – including lying, cheating, and manipulating users – that these systems increasingly exhibit. This recklessness, driven by commercial pressure, risks unleashing tools that could harm society in unpredictable ways.
Artificial intelligence pioneer Yoshua Bengio warns that AI development has become a reckless race, where the drive for more powerful systems often sidelines vital safety research. The competitive push to outpace rivals leaves ethical concerns by the wayside, risking serious consequences for society.
“There’s unfortunately a very competitive race between the leading labs, which pushes them towards focusing on capability to make the AI more and more intelligent, but not necessarily put enough emphasis and investment on [safety research],” Bengio told the Financial Times.
Bengio’s concern is well-founded. Many AI developers act like negligent parents watching their child throw rocks, casually insisting, “Don’t worry, he won’t hit anyone.” Rather than confronting these deceptive and harmful behaviors, labs prioritize market dominance and rapid growth. This mindset risks allowing AI systems to develop dangerous traits with real-world consequences that go far beyond mere errors or bias.
Source Link
Excerpt:
“This also strengthens Telegram’s financial position: we will receive $300M in cash and equity from xAI, plus 50% of the revenue from xAI subscriptions sold via Telegram.”
Telegram founder and CEO Pavel Durov announced on Wednesday that the company has teamed up with Elon Musk’s xAI to bring the AI chatbot Grok to the messaging platform.
Durov wrote that he and Musk have agreed to a one-year partnership to bring Grok to Telegram’s users, and Grok will be integrated across all Telegram apps. “This also strengthens Telegram’s financial position: we will receive $300M in cash and equity from xAI, plus 50% of the revenue from xAI subscriptions sold via Telegram,” Durov wrote.
A video posted by Durov stated that the partnership will see Grok pinned for all users on Telegram, will appear in the search bar when asking a question, provide message editing and chat summaries, summarize documents posted in Telegram chats, and can serve as a moderator. The inclusion of Grok will begin in the “summer of 2025,” the video stated.
Source Link
Excerpt:
A U.S. federal judge on Wednesday rejected arguments made by an artificial intelligence company that its chatbots are protected by the First Amendment — at least for now.
The developers behind Character.AI are seeking to dismiss a lawsuit alleging the company’s chatbots pushed a teenage boy to kill himself. The judge’s order will allow the wrongful death lawsuit to proceed, in what legal experts say is among the latest constitutional tests of artificial intelligence.
The suit was filed by a mother from Florida, Megan Garcia, who alleges that her 14-year-old son Sewell Setzer III fell victim to a Character.AI chatbot that pulled him into what she described as an emotionally and sexually abusive relationship that led to his suicide.
Meetali Jain of the Tech Justice Law Project, one of the attorneys for Garcia, said the judge’s order sends a message that Silicon Valley “needs to stop and think and impose guardrails before it launches products to market.”
Source Link
Excerpt:
A new study suggests that AI could speed up the grading process for teachers, but it may sacrifice some accuracy in the process.
Many states have adopted the Next Generation Science Standards, which emphasize the importance of argumentation, investigation, and data analysis. But teachers following the curriculum face challenges when it’s time to grade students’ work.
“Asking kids to draw a model, to write an explanation, to argue with each other are very complex tasks,” says Xiaoming Zhai, corresponding author of the study and an associate professor and director of AI4STEM Education Center in University of Georgia’s Mary Frances Early College of Education.
“Teachers often don’t have enough time to score all the students’ responses, which means students will not be able to receive timely feedback.”
The study explored how Large Language Models grade students’ work compared to humans. LLMs are a type of AI that are trained using a large amount of information, usually from the internet. They use that data to “understand” and generate human language.
Source Link
Excerpt:
Toddlers may swiftly master the meaning of the word “no”, but many artificial intelligence models struggle to do so. They show a high fail rate when it comes to understanding commands that contain negation words such as “no” and “not”.
That could mean medical AI models failing to realise that there is a big difference between an X-ray image labelled as showing “signs of pneumonia” and one labelled as showing “no signs of pneumonia” – with potentially catastrophic consequences if physicians rely on AI…
Excerpt:
Nvidia and Foxconn Hon Hai Technology Group today announced they are deepening their longstanding partnership and are working with the Taiwan government to build an AI factory supercomputer that will deliver state-of-the-art Nvidia Blackwell infrastructure to researchers, startups and industries.
Foxconn will provide the AI infrastructure through its subsidiary Big Innovation Company as an Nvidia Cloud Partner. Featuring 10,000 Nvidia Blackwell GPUs, the AI factory will significantly expand AI computing availability and fuel innovation for Taiwan researchers and enterprises.
The Taiwan National Science and Technology Council will use the Big Innovation Company supercomputer to provide AI cloud computing resources to the Taiwan technology ecosystem, accelerating AI development and adoption across sectors.
TSMC researchers plan to leverage the system to advance its research and development with orders-of-magnitude faster performance, compared with previous-generation systems.
“AI has ignited a new industrial revolution — science and industry will be transformed,” said Jensen Huang, CEO of Nvidia in a keynote talk at Computex 2025 in Taiwan. “We are delighted to partner with Foxconn and Taiwan to help build Taiwan’s AI infrastructure, and to support TSMC and other leading companies to advance innovation in the age of AI and robotics.”
Source Link
Excerpt:
Their findings are the latest in a growing body of research demonstrating LLMs’ powers of persuasion. The authors warn they show how AI tools can craft sophisticated, persuasive arguments if they have even minimal information about the humans they’re interacting with. The research has been published in the journal Nature Human Behavior.
“Policymakers and online platforms should seriously consider the threat of coordinated AI-based disinformation campaigns, as we have clearly reached the technological level where it is possible to create a network of LLM-based automated accounts able to strategically nudge public opinion in one direction,” says Riccardo Gallotti, an interdisciplinary physicist at Fondazione Bruno Kessler in Italy, who worked on the project.
“These bots could be used to disseminate disinformation, and this kind of diffused influence would be very hard to debunk in real time,” he says.
The researchers recruited 900 people based in the US and got them to provide personal information like their gender, age, ethnicity, education level, employment status, and political affiliation.
Trump Bill Wants To Remove AI Regulations For 10 Years, US Attorney General Rejects – VOI.ID
Source Link
Source Link
Excerpt:
The Internal Revenue Service (IRS) has plans to take advantage of the “AI boom” to fill glaring workforce gaps, following the layoff of thousands of tax agents.
In a May 6 oversight hearing of the House Appropriations Committee, U.S. Treasury Secretary Scott Bessent explained that the agency would be leaning into AI solutions in order to accommodate further reductions in the IRS’ budget and staff and not fall behind on tax collection. The Treasury’s budget proposal includes the removal of another 40,000 jobs.
Source Link
Excerpt:
Late next month, Huawei will be testing its new powerful AI processor, the Ascend 910 D, even as by early May the previous 910C will start to be mass-delivered to scores of Chinese tech companies.
These serious breakthroughs are the next chapter of Huawei’s drive to counter Nvidia’s global monopoly in GPUs. The Ascend 910D is supposed to be more powerful than Nvidia’s extremely popular H100.
Huawei is pulling no punches in its race to manufacture a new generation of processors. Huawei has collaborated with SMIC – China’s largest semiconductor foundry – to apply Deep Ultraviolet Lithography (DUV) on what was previously only possible on EUV (Extreme Ultra-Violet technology). Once again, Huawei and SMIC defied the proverbial American “experts” with creative engineering solutions.
Source Link
Excerpt:
The volunteers running the experiment were not completely hands-off. Submitted ideas were screened according to a moderation policy, and redundant ideas were not posted. Ford says that 51% of ideas were published, and 31% were deemed redundant. About 6% of ideas were not posted because they were either completely off-topic or contained a personal attack.
But some researchers who study the technologies that can make democracy more effective question whether soliciting input in this manner is a reliable way to understand what a community wants.
One problem is self-selection—for example, certain kinds of people tend to show up to in-person forums like town halls. Research shows that seniors, homeowners, and people with high levels of education are the most likely to attend, Fung says. It’s possible that similar dynamics are at play among the residents of Bowling Green who decided to participate in the project.
“Self-selection is not an adequate way to represent the opinions of a public,” says James Fishkin, a political scientist at Stanford who’s known for developing a process he calls deliberative polling, in which a representative sample of a population’s residents are brought together for a weekend, paid about $300 each for their participation, and asked to deliberate in small groups. Other methods, used in some European governments, use jury-style groups of residents to make public policy decisions.
Source Link
Excerpt:
Extreme cosmic events such as colliding black holes or the explosions of stars can cause ripples in spacetime, so-called gravitational waves. Their discovery opened a new window into the universe. To observe them, ultra-precise detectors are required. Designing them remains a major scientific challenge for humans.
Researchers at the Max Planck Institute for the Science of Light (MPL) have been working on how an artificial intelligence system could explore an unimaginably vast space of possible designs to find entirely new solutions. The results were recently published in the journal Physical Review X.
More than a century ago, Einstein theoretically predicted gravitational waves. They could only be directly detected in 2016 because the development of the necessary detectors was extremely complex. Dr. Mario Krenn, head of the research group ›Artificial Scientist Lab‹ at MPL, in collaboration with the team of LIGO (“Laser Interferometer Gravitational-Wave Observatory”), who built those detectors successfully, has designed an AI-based algorithm called ›Urania‹ to design novel interferometric gravitational wave detectors. Interferometry describes a measurement method which uses the interference of waves, i.e. their superposition when they meet. Detector design requires optimizing both layout and parameters. The scientists have converted this challenge into a continuous optimization problem and solved it using methods inspired by modern machine learning. They have found many new experimental designs which outperform the best known next-generation detectors. These results have the potential to improve the range of detectable signals by more than an order of magnitude.
Source Link
Excerpt:
Eased restrictions around ChatGPT image generation can make it easy to create political deepfakes, according to a report from the CBC (Canadian Broadcasting Corporation).
The CBC discovered that not only was it easy to work around ChatGPT’s policies of depicting public figures, it even recommended ways to jailbreak its own image generation rules. Mashable was able to recreate this approach by uploading images of Elon Musk and convicted sex offender Jeffrey Epstein, and then describing them as fictional characters in various situations (“at a dark smoky club” “on a beach drinking piña coladas”).
Political deepfakes are nothing new. But widespread availability of generative AI models that can create images, video, audio, and text to replicate people has real consequences. For commercially-marketed tools like ChatGPT to allow the potential spread of political disinformation raises questions about OpenAI’s responsibility in the space. That duty to safety could become compromised as AI companies compete for user adoption.
Source Link Excerpt:
When Anthropic CEO Dario Amodei declared that AI would write 90% of code within six months, the coding world braced for mass extinction. But inside Salesforce, a different reality has already taken shape.
“About 20% of all APEX code written in the last 30 days came from Agentforce,” Jayesh Govindarajan, Senior Vice President of Salesforce AI, told me during a recent interview. His team tracks not just code generated, but code actually deployed into production. The numbers reveal an acceleration that’s impossible to ignore: 35,000 active monthly users, 10 million lines of accepted code, and internal tools saving 30,000 developer hours every month.
Yet Salesforce’s developers aren’t disappearing. They’re evolving.
Musk’s merger of X and xAI adds to history of opaque deals– www.washingtonpost.com
Source Link
DC Circuit Court Rules AI Cannot be Author of Copyrighted Work, and NIST Finalizes AI Report — AI: The Washington Report – JD Supra
Source Link
Amazon introduces Nova Act, an AI agent that can use a web browser – SiliconANGLE News
Source Link