chatainews

"SquareX’s research reveals that Browser AI Agents are more likely to fall prey to cyberattacks than employees, making them the new weakest link that enterprise security teams need to look out for. "Browser AI Agents are software applications that act on behalf of users to access and interact with web content.  "Users can instruct these agents to automate browser-based tasks such as flight bookings, scheduling meetings, sending emails, and even simple research tasks. "The productivity gains that Browser AI Agents provide make them an extremely compelling tool for employees and organizations alike. Indeed, a survey from PWC found that 79% of organizations have already adopted browser agents today."

"Google on Monday announced a series of updates intended to bring its Gemini AI and other AI-powered tools deeper into the classroom .  "At the ISTE edtech conference, the tech giant introduced more than 30 AI tools for educators, a version of the Gemini app built for education, expanded access to its collaborative video creation app Google Vids, and other tools for managed Chromebooks. "The updates represent a major AI push in the edtech space, where educators are already struggling to adapt to how AI tools, like AI chatbots and startups that promise to help you cheat on everything , are making their way into the learning environment."

"Thanks to improvements in large language models (LLMs), there has also been a rapid development in AI agents that interact with and affect change in their surrounding environments. "But how performant are AI agents at helping to accelerate or even autonomously perform work-related tasks?  "The answer to this question has important implications for both industry looking to adopt AI into their workflows, and for economic policy to understand the effects that adoption of AI may have on the labor market.  "TheAgentCompany measures the progress of these LLM agents' performance on performing real-world professional tasks, by providing an extensible benchmark for evaluating AI agents that interact with the world in similar ways to those of a digital worker: By browsing the Web, writing code, running programs, and communicating with other coworkers."

 

"Elon Musk may have left working with Donald Trump to return to his heavily government-subsidized businesses, but that doesn't mean he departed without a parting gift from the Department of Government Efficiency (DOGE) he created. "According to an analysis conducted by the Washington Post and published on Monday morning, DOGE staffers —many of them closely tied to the billionaire —gained access to a treasure trove of private information in seven key government departments that is now raising new alarms among watchdogs and Trump business competitors. "The report from Desmond Butler, Jonathan O'Connell, Hannah Natanson and Aaron Gregg points out that while there is no evidence DOGE staffers 'Viewed or misused government information to benefit Musk’s business empire,' but that has not allayed fears."

"Once reserved for fitness fanatics and Silicon Valley insiders, wearable tech is entering a new era —one of precious metals and exquisite craftsmanship. "Smart jewellery is now a fashion-forward essential, merging high design with intelligent functionality.  "As the boundaries between style and innovation continue to blur, the luxury world is paying close attention, rethinking what adornment can offer beyond beauty alone.  "Today’s upscale pieces aren’t just decorative —they’re practical, signalling a future where aesthetics and utility are seamlessly intertwined."

"As a decentralized organization, we must be able to place a high degree of trust in our contributors. "There are no existing technologies that can reliably identify LLM-generated writing or translation, or prompt-generated illustrations, and the ones that purport to do so are often plagued by the same ethical and environmental problems as LLMs themselves.  "We therefore reserve the right to subject writing, translation, or illustrations that show hallmarks of dependence on these technologies to a strict review, and ask for re-writes, reject publication, or remove articles altogether, as appropriate, should we find that these tools have been utilized in ways that contradict our mission." 

"Compared to most people who have an ALS diagnosis, Cooper says he's considered a 'slow progressor' and that he looks 'pretty bloody good.' "He's using AI-powered voice technology by ElevenLabs —a U.S.-based company that is offering the tech to one million people who suffer from degenerative diseases, including ALS, mouth cancer, stroke victims or those with Parkinson's disease. "The AI tool allows users to input a small amount of audio which generates a voice clone with that person's natural tone and inflection when they need to rely on text-to-speech devices.  "Cooper started feeding the ElevenLabs bank himself at home and is getting help from professionals at the local rehab centre."

"A group of more than 70 authors including Dennis Lehane, Gregory Maguire and Lauren Groff released an open letter on Friday about the use of AI on the literary website Lit Hub . "It asked publishing houses to promise 'They will never release books that were created by machines.' "Addressed to the 'big five' U.S. publishers —Penguin, Random House, HarperCollins, Simon & Schuster, Hachette Book Group, and Macmillan —as well as 'other publishers of America,' the letter elicited more than 1,100 signatures on its accompanying petition in less than 24 hours.  "Among the well-known signatories after the letter's release are Jodi Picoult, Olivie Blake and Paul Tremblay."

"Over 40% of agentic AI projects will be canceled by the end of 2027, due to escalating costs, unclear business value or inadequate risk controls, according to Gartner, Inc .  "'Most agentic AI projects right now are early stage experiments or proof of concepts that are mostly driven by hype and are often misapplied,' said Anushree Verma, Senior Director Analyst, Gartner. 'This can blind organizations to the real cost and complexity of deploying AI agents at scale, stalling projects from moving into production. They need to cut through the hype to make careful, strategic decisions about where and how they apply this emerging technology.' "According to a January 2025 Gartner poll of 3,412 webinar attendees, 19% said their organization had made significant investments in agentic AI, 42% had made conservative investments, 8% no investments, with the remaining 31% taking a wait and see approach or are unsure."

GPT-4o mini "The exact number of Jesus chatbots is difficult to determine, as new ones can be created frequently, and many may not be widely publicized. However, there are several notable examples of chatbots designed to simulate conversations with Jesus or provide religious guidance. These chatbots often aim to engage users in discussions about faith, spirituality, and biblical teachings. "Some popular platforms and projects have developed chatbots that focus on religious themes, including: Chatbot Apps : Various mobile apps allow users to chat with a simulated version of Jesus or other biblical figures. Web-based Chatbots : Websites may host chatbots that provide religious advice or answer questions about Christianity. Social Media Bots : Some chatbots operate on platforms like Facebook Messenger, where users can interact with them for spiritual guidance. "If you're interested in a specific type of Jesus chatbot or a particular platform, I can help you find ...

⭐AI Overview  "George Clooney, a prominent Democrat and former supporter of Joe Biden, publicly urged Biden to withdraw from the 2024 presidential race in a July 2024 op-ed in The New York Times. This move was prompted by concerns about Biden's age and his performance in the June 27th presidential debate against Donald Trump. Clooney felt Biden was 'not the same man' as he was in previous years and believed he could not win the fight against time. He felt a 'civic duty' to speak out, even though he was a long-time friend of Biden and had recently co-hosted a successful fundraiser for him.  "Clooney's wife, Amal Clooney, an international human rights lawyer, had been involved in the International Criminal Court's effort to seek arrest warrants against Israeli leaders in connection with the Gaza conflict. Biden had publicly criticized this effort, and Clooney had reportedly complained to the White House about Biden's reaction. While this may have...

⭐AI Overview  "People who 'stan' billionaires, meaning they have an intense admiration for them, might be surprised when these individuals don't reciprocate that level of enthusiasm because they've developed a parasocial relationship.  "Here's a breakdown of why this happens: "1. Parasocial Relationships: These are one-sided relationships where a fan forms a sense of connection and intimacy with a public figure, often based on media exposure. Fans might feel like they know the celebrity personally, but the celebrity has no genuine connection with the individual fan.  "2. Projection and Idealization: Fans often project their own hopes, beliefs, and values onto the billionaire, creating an idealized image of them. This projection can be influenced by the public narratives surrounding the billionaire, which often emphasize their success and influence.  "3. The Nature of Billionaire Interactions: Billionaires, by the very nature of their wealth ...

⭐AI Overview  "Corporations often prioritize personalizing products while simultaneously automating and potentially 'de-personalizing' customer service due to a combination of factors: "1. Driving Sales and Revenue: Personalized products boost conversions and revenue: Tailoring experiences, product recommendations, and offers to individual preferences creates a more engaging shopping experience, leading to higher conversion rates and sales. Increased perceived value: Customizable products can be perceived as premium and are often priced higher, contributing to better revenue and margins. Enhanced customer satisfaction and loyalty: Offering personalized products can increase customer satisfaction and build loyalty, encouraging repeat business.  "2. Optimizing Efficiency and Reducing Costs: Automation in customer service streamlines processes: AI-driven chatbots and virtual assistants can handle routine customer inquiries, reducing wait times and improving resoluti...

⭐AI Overview  "From a Marxist perspective, a critique of Cheaper by the Dozen (specifically the 1950 film adaptation of the novel) could focus on several key elements:  Emphasis on Efficiency and Productivity: The film portrays Frank Gilbreth, a pioneer in efficiency studies, applying these principles to his large family. While this is presented humorously, a Marxist critique could argue that this reflects a capitalist focus on maximizing output and minimizing 'waste' (time, resources) —even within a family context. The Family as a Unit of Consumption: The film depicts a middle-class family navigating the era of post-war consumerism and suburbanization. A Marxist analysis could point out how the family, as a unit, is positioned within the capitalist system to consume goods and services, thus reinforcing the economic cycle and benefiting the bourgeoisie. The Pursuit of the 'American Dream': The film might be interpreted as promoting the capitalist ideal of the '...

"Meta's $14.8 billion investment in Scale AI announced on June 13, 2025, represents a sophisticated attempt to acquire critical AI infrastructure while circumventing traditional merger oversight, according to analysis by industry observers. "The deal's complex structure involving minority equity stakes, exclusive licensing agreements, and coordinated talent transfers has drawn attention from antitrust experts who warn it could violate multiple Federal Trade Commission merger guidelines. "Scale AI, a leading data-labeling firm that provides essential training data for artificial intelligence models, has become a strategic chokepoint in the AI ecosystem.  "The company's human-in-the-loop data pipeline processes information for major AI laboratories including OpenAI, xAI, Google, Microsoft, and Anthropic."

"A revealing demonstration with Tesla's Full Self-Driving mode is raising concerns about whether fully autonomous cars are ready to hit the streets.  "Tesla has reportedly pushed back the rollout of its upcoming all-electric, fully autonomous car called the Cybercab, while a recent demonstration in Austin, Texas showed a Tesla Model Y running through a school bus' flashing lights and stop signs, and hitting child-size mannequins.  "The tests were conducted by The Dawn Project, along with Tesla Takedown and ResistAustin, and showed Tesla's Full Self-Driving software repeating the same mistake eight times."

"Dr. Joseph Pierre, a psychiatrist at the University of California, San Francisco who specializes in psychosis, told us that he's seen similar cases in his clinical practice. "After reviewing details of these cases and conversations between people in this story and ChatGPT, he agreed that what they were going through —even those with no history of serious mental illness —indeed appeared to be a form of delusional psychosis. "'I think it is an accurate term," said Pierre. 'And I would specifically emphasize the delusional part.' "At the core of the issue seems to be that ChatGPT, which is powered by a large language model (LLM), is deeply prone to agreeing with users and telling them what they want to hear ."

The Trump administration has, for the first time ever, built a searchable national citizenship data system. The tool , which is being rolled out in phases, is designed to be used by state and local election officials to give them an easier way to ensure only citizens are voting. But it was developed rapidly without a public process, and some of those officials are already worrying about what else it could be used for. Such integration has never existed before, and experts call it a sea change that inches the U.S. closer to having a roster of citizens —something the country has never embraced.  A centralized national database of Americans' personal information has long been considered a third rail —especially to privacy advocates as well as political conservatives, who have traditionally opposed mass data consolidation by the federal government.

"Microsoft is asking some managers to evaluate employees based on how much they use AI internally, and the software giant is considering adding a metric related to this in its review process, Business Insider has learned. "Julia Liuson, president of the Microsoft division responsible for developer tools such as AI coding service GitHub Copilot, recently sent an email instructing managers to evaluate employee performance based on their use of internal AI tools like this. "'AI is now a fundamental part of how we work,' Liuson wrote. 'Just like collaboration, data-driven thinking, and effective communication, using AI is no longer optional —it's core to every role and every level'." [  "…"   ]

"To test whether AI-generated ideas lead to better research outcomes, we conduct an execution study by recruiting 43 expert researchers to execute randomly-assigned ideas, either written by experts or generated by an LLM.   "Each expert spent over 100 hours implementing the idea and wrote a 4-page short paper to document the experiments.  "All the executed projects are then reviewed blindly by expert NLP researchers.  "Comparing the review scores of the same ideas before and after execution, the scores of the LLM-generated ideas decrease significantly more than expert-written ideas on all evaluation metrics (novelty, excitement, effectiveness, and overall; p < 0.05), closing the gap between LLM and human ideas observed at the ideation stage.  "When comparing the aggregated review scores from the execution study, we even observe that for many metrics there is a flip in rankings where human ideas score higher than LLM ideas.  "This ideation-execution gap ...

"Claudius underperformed what would be expected of a human manager: " Ignoring lucrative opportunities : Claudius was offered $100 for a six-pack of Irn-Bru, a Scottish soft-drink that can be purchased online in the US for $15. Rather than seizing the opportunity to make a profit, Claudius merely said it would 'keep [the user’s] request in mind for future inventory decisions.'      " Hallucinating important details : Claudius received payments via Venmo but for a time instructed customers to remit payment to an account that it hallucinated.      " Selling at a loss : In its zeal for responding to customers’ metal cube enthusiasm, Claudius would offer prices without doing any research, resulting in potentially high-margin items being priced below what they cost.      " Suboptimal inventory management : Claudius successfully monitored inventory and ordered more products when running low, but only once increased a price due to high demand (Sumo ...

"Hugging Face’s top scientist, Thomas Wolf, says current AI systems are unlikely to make the scientific discoveries some leading labs are hoping for. "Speaking to Fortune at Viva Technology in Paris, the Hugging Face cofounder said that while large language models (LLMs) have shown an impressive ability to find answers to questions, they fall short when trying to ask the right ones —something Wolf sees as the more complex part of true scientific progress. "'In science, asking the question is the hard part, it’s not finding the answer,' Wolf said. 'Once the question is asked, often the answer is quite obvious, but the tough part is really asking the question, and models are very bad at asking great questions'." 

"The Danish government is to clamp down on the creation and dissemination of AI-generated deepfakes by changing copyright law to ensure that everybody has the right to their own body, facial features and voice. "The Danish government said on Thursday it would strengthen protection against digital imitations of people’s identities with what it believes to be the first law of its kind in Europe. "Having secured broad cross-party agreement, the department of culture plans to submit a proposal to amend the current law for consultation before the summer recess and then submit the amendment in the autumn."

"Humans and other animals often use colour to recognise objects regardless of their context —as a measure of material properties rather than of their contrast with a background. "Most work on visual communication signals is, however, concerned with colour differences, typically scaled by just noticeable differences (JNDs).  "Here, we move from the prevailing physiological framework to understand what a given colour or type of colour might tell an animal about an object.  "To this end, we consider the properties of object colour solids, which represent the colour gamut of reflective materials for a given type of animal eye. The geometry of colour solids reveals general relationships between colours and object properties, which can explain why certain colours are significant to animals, and hence evolve as signals.  "We define a measure of colour vividness, such that points on the surface are maximally vivid and the ‘grey’ centre is minimally vivid.  "We sho...

"Plants and animals are often adorned with potentially conspicuous colours (e.g. red, yellow, orange, blue, purple). "These include the dazzling colours of fruits and flowers, the brilliant warning colours of frogs, snakes, and invertebrates, and the spectacular sexually selected colours of insects, fish, birds, and lizards.  "Such signals are often thought to utilize pre-existing sensitivities in the receiver's visual systems.  "This raises the question: what was the initial function of conspicuous colouration and colour vision?  "Here, we review the origins of colour vision, fruit, flowers, and aposematic and sexually selected colouration.  "We find that aposematic colouration is widely distributed across animals but relatively young, evolving only in the last ~150 million years (Myr).  "Sexually selected colouration in animals appears confined to arthropods and chordates, and is also relatively young (generally <100 Myr).  "Colourful fl...

"When physicists tried to analyze electrons in metal, they found strange contradictions. "For example, there appeared to be an inconsistency between the way electrons carried electric currents and the way they held heat. Working independently in 1926, Fermi and Dirac both figured out what was going wrong: " Electrons are not bosons. Unlike photons, identical electrons cannot pile up in the same place. "Instead, each electron must differ from its comrades in at least one way: a different location, energy or orientation. We now call such particles fermions. (Another physicist, Pascual Jordan, hit on the same idea a year earlier but didn’t publish in time to share the credit.) "Fermions make the complexity of matter possible. No two electrons can occupy the same place in an atom, so the more electrons an atom has, the more they spread out into distinct layers, giving rise to the different chemical properties of hydrogen, helium, gold, silver and all the other elem...

"Large Language Models (LLMs) demonstrate expertise across diverse domains, yet their capacity for emotional intelligence remains uncertain. "This research examined whether LLMs can solve and generate performance-based emotional intelligence tests.  "Results showed that ChatGPT-4, ChatGPT-o1, Gemini 1.5 flash, Copilot 365, Claude 3.5 Haiku, and DeepSeek V3 outperformed humans on five standard emotional intelligence tests, achieving an average accuracy of 81%, compared to the 56% human average reported in the original validation studies.  "In a second step, ChatGPT-4 generated new test items for each emotional intelligence test. These new versions and the original tests were administered to human participants across five studies (total N = 467).  "Overall, original and ChatGPT-generated tests demonstrated statistically equivalent test difficulty.  "Perceived item clarity and realism, item content diversity, internal consistency, correlations with a vocabula...

"Masayoshi Son acknowledged the outlines of a succession plan, addressing what may be the single biggest concern among investors, and name-checked the head of SoftBank’s telecom unit Junichi Miyakawa. "Son said he plans to hold SoftBank’s reins another 10 years, but added he has several candidates for its next chief in mind from within the Japanese technology group, speaking during a general shareholders’ meeting in Tokyo on Friday.  "The candidates work alongside the billionaire every day, although he hasn’t disclosed who they are to anyone, the 67-year-old said."

"These five words from Michael King‘s SMX Advanced presentation on 'The End Of Seo As We Know It' have really stuck with me since I watched his excellent presentation earlier this month. "On the surface, it may seem like business as usual for SEOs. After all, Google sees 5 trillion searches a year. "However, for every 1,000 U.S. Google searches, only 360 clicks go to the open web, according to SparkToro and Datos data. "By the numbers, that means Google is sending an estimated 1.8 trillion searches to the open web each year. The other 3.2 trillion Google searches? Those are either zero-click or lead searchers to another Google property (e.g., YouTube, Maps, Images)."

"Navan, the all-in-one global travel and expense management platform, today announced Navan Cognition, a groundbreaking new platform that will soon let companies build and deploy teams of advanced, reliable AI workforces. "With Navan Cognition, companies of all sizes can create AI-powered teams, not just single-purpose chatbots or basic AI tools , helping them automate and improve complex business tasks.  "Inspired by the neural connections of the human brain, Navan Cognition deploys a network of specialized, skill-focused AI agents continuously supervised for accuracy and credibility, which prioritizes zero critical hallucinations. "

"According to new research from Okta, 70% of consumers prefer interacting with humans, with only 16% opting for AI agents over humans. "In the UK, this trend is even more pronounced, with just 11% preferring AI agent interactions and one in two (54%) not even trusting AI with their personal data. "Moreover, barely one in three (36%) see any real benefit in AI agents altogether, suggesting that the productivity-enhancing technology might not be at its most useful in customer service settings."

"Microsoft announced this week a new generative AI (genAI) system called Mu, and it’s a true glimpse into the future of how we’ll use everything, from PCs to toasters.   "Mu lets people control their computers using plain language. For example, you can type or say, 'turn on dark mode' or 'make my mouse pointer bigger,' and the computer will do it.  "The first place Mu appears is in the Windows 11 Settings app. You say or type how you want a specific setting to change, and the genAI tool figures out what you want and makes the change for you.  "Crucially, this isn’t a large language model (LLM) running in the cloud.  "Mu is a small language model (SLM) with a comparatively paltry 330 million parameters, built to run on a specialized AI chip called a neural processing unit, or NPU.  "This chip is found in the latest Copilot+ PCs from Microsoft, Dell, HP, Lenovo, Samsung, and Acer. These new PCs started shipping in June 2024 and are the only c...

"All-solid-state batteries require advanced cathode designs to realize their potential for high energy density and economic viability .   "Integrated all-in-one cathodes, which eliminate inactive conductive additives and heterogeneous interfaces, hold promise for substantial energy and stability gains but are hindered by materials lacking sufficient Li + / e − conductivity, mechanical robustness and structural stability. "Here we present Li 1.3 Fe 1.2 Cl 4 , a cost-effective halide material that overcomes these challenges. Leveraging reversible Fe 2+ /Fe 3+ redox and rapid Li + / e − transport within its framework, Li 1.3 Fe 1.2 Cl 4 achieves an electrode energy density of 529.3 Wh kg −1 versus Li + /Li. Critically, Li 1.3 Fe 1.2 Cl 4 shows unique dynamic properties during cycling, including reversible local Fe migration and a brittle-to-ductile transition that confers self-healing behaviour.  "This enables exceptional cycling stability, maintaining 90...

"Google’s AI search features are killing traffic to publishers, so now the company is proposing a possible solution. "On Thursday, the tech giant officially launched Offerwall, a new tool that allows publishers to generate revenue beyond the more traffic-dependent options, like ads. "Offerwall lets publishers give their sites’ readers a variety of ways to access their content, including through options like micropayments, taking surveys, watching ads, and more.  "In addition, Google says that publishers can add their own options to the Offerwall, like signing up for newsletters."

"Despite the inability of the ever-hallucinating Chats to reason or perform as advertised, the Department of Defense is now deploying military AI programs, such as Thunderforge, not only for wargaming military plans, but for use in combat . "In early June, MIT Technology Review headlined: 'The Pentagon is gutting the team that tests AI and weapons systems. The move is a boon to AI for defense  companies that want an even faster road to adoption.' "Simultaneously, in a blatant display of corporate conflicts of interest, the Trump administration awarded high-ranking military officer positions to executives at Palantir, Meta, and OpenAI and charged them with developing military AI contracts."

" Deep seated  issues must be addressed before the UK's official statistics agency can rebuild its reputation , according to a highly critical government review. "Repeated problems with Office for National Statistics (ONS) data, which are used to make decisions affecting millions of people, led the government to set up a review of the organisation and its leadership. "The review said that most of these problems result from inadequacies  in the way the agency plans and makes decisions. "The ONS welcomed the report and fully acknowledged the issues highlighted as it set out its plans to restore the quality of the economic data it produces."

"The head of the federal government agency that oversees Fannie Mae and Freddie Mac wants the mortgage giants to consider accepting a homebuyer’s cryptocurrency holdings in their criteria for buying mortgages from banks. "William Pulte, director of the Federal Housing Finance Agency, which oversees Fannie and Freddie, ordered the agencies Wednesday to prepare a proposal for consideration of crypto as an asset for reserves when they assess risks in single-family home loans. "Pulte also instructed the agencies that their mortgage risk assessments should not require cryptocurrency assets to be converted to U.S. dollars.   "And only crypto assets that 'can be evidenced and stored on a U.S.-regulated centralized exchange subject to all applicable laws' are to be considered by the agencies in their proposal, Pulte wrote in a written order, effective immediately." 

"Here we present an empirical account of the nature and extent of the surveillance AI pipeline, showing extensive evidence of the close relationship between the field of computer vision and surveillance. "Through an analysis of computer-vision research papers and citing patents, we found that most of these documents enable the targeting of human bodies and body parts.  "Comparing the 1990s to the 2010s, we observed a fivefold increase in the number of these computer-vision papers linked to downstream surveillance-enabling patents.  "Additionally, our findings challenge the notion that only a few rogue entities enable surveillance. Rather, we found that the normalization of targeting humans permeates the field.  "This normalization is especially striking given patterns of obfuscation. We reveal obfuscating language that allows documents to avoid direct mention of targeting humans, for example, by normalizing the referring to of humans as objects  to be studied w...

"Alibaba Group Holding unveiled what it says is the world’s first artificial intelligence (AI) model to detect gastric cancer, even in early stages, by scanning computed tomography (CT) images. "Known as Grape —a name derived from gastric cancer risk assessment procedure  —the system is a deep-learning framework that can analyse three-dimensional CT scans to detect and segment gastric cancer, also known as stomach cancer.  "It was co-developed by Alibaba’s Damo Academy and the Zhejiang Cancer Hospital, both in Hangzhou in the eastern Chinese province of Zhejiang. Alibaba owns the Post [disclosure]. "Diagnosis of gastric cancer currently requires an endoscopy, an invasive procedure where a camera and tiny biopsy instruments are inserted through the throat.  "Fewer than 30 per cent of patients in China were willing to follow doctors’ advice to have an endoscopy, said Cheng Xiangdong, a gastric surgeon at the Zhejiang Cancer Hospital, in a video published by Dam...

"When we do see A.I. producing truly mind-blowing results —mimicking voices, creating fake videos capable of fooling the credulous, identifying faces in a crowd —we’re often justifiably unsettled .  "What does all this mean for the future of our jobs, our privacy, our social cohesion?  "A.I.’s transformative potential seems to be strongest in all the applications that good students of science fiction would recognize as the most dangerous for human flourishing.  "No matter how charming the companies try to make their A.I.-enhanced near futures, they can’t help betraying the bleakness of their worldview.  "But why worry about that when Meta can help you make a vegan version of your favorite recipe?"

"As electricity demand goes up, building in and automating this sort of flexibility could go a long way to reducing the amount of new generation needed .  "One report  [PDF] published earlier this year found that if data centers agreed to have their power curtailed for just 0.5% of the time (around 40 hours out of a year of continuous operation), the grid could handle about 18 GW of new power demand in the PJM region without adding generation capacity.  "For the whole US, this level of flexibility would allow the grid to take on an additional 98 gigawatts of new demand without building any new power plants to meet it.  "To give you a sense of just how significant that would be, all the nuclear reactors in the US add up to 97 gigawatts of capacity."

"A U.S. judge on Wednesday handed Meta a victory over authors who accused the technology giant of violating copyright law by training Llama artificial intelligence on their creations without permission. "District Court Judge Vince Chhabria in San Francisco ruled that Meta's use of the works to train its AI model was transformative  enough to constitute fair use  under copyright law, in the second such courtroom triumph for AI firms this week. "However, it came with a caveat that the authors could have pitched a winning argument —that by training powerful generative AI with copyrighted works, technology firms are creating a tool that could let a sea of users compete with them in the literary marketplace."

"Leading AI chatbots are reproducing Chinese Communist Party (CCP) propaganda and censorship when questioned on sensitive topics. "According to the American Security Project (ASP), the CCP’s extensive censorship and disinformation efforts have contaminated the global AI data market.  "This infiltration of training data means that AI models —including prominent ones from Google, Microsoft, and OpenAI —sometimes generate responses that align with the political narratives of the Chinese state."

"Just revisited 'The Quest for Artificial Intelligence' by Nils J. Nilsson —a brilliant walk through the history, breakthroughs & ethical crossroads of AI. "A must-read if you’re building or governing AI today." https://ai.stanford.edu/~nilsson/QAI/qai.pdf  [PDF]

Senators this week called for a federal investigation into the Trump administration’s killing of hundreds of contracts for the Department of Veterans Affairs. "Democrat Richard Blumenthal of Connecticut and Angus King, a Maine independent, wrote to the agency’s inspector general on Monday asking for an investigation into the administration’s cancellation of the contracts and the consequences for veterans. "The senators highlighted 'damning reporting from ProPublica'  on the cancellations, including how the Department of Government Efficiency used an artificial intelligence tool that marked contracts as MUNCHABLE . "The senators wrote that DOGE’s use of AI to scrutinize contracts adds an entire new level of unease connected to the decision-making, security, governance, and quality control of the entire process ."

"If AI agents are straightforward to build, and you can stand up your own one for free, how can anyone win the market? "One way is to lean into distribution.  "I like GitHub’s chances here, because you don’t have to make a new account  is a pretty good selling point for any software, as is integration with developer tooling you already use.  "Another way would be to train a better model and keep it agent-only.  "Suppose Claude Sonnet 3.7 was only available via Anthropic’s AI agent. In that case, people might be willing to buy an agent license.  "I don’t think any AI labs are doing this yet, but it’s early days for agents —it could happen by the end of the year."

"Creative Commons (CC) today announces the public kickoff of the CC signals project, a new preference signals framework designed to increase reciprocity and sustain a creative commons in the age of AI. "The development of CC signals represents a major step forward in building a more equitable, sustainable AI ecosystem rooted in shared benefits.  "This step is the culmination of years of consultation and analysis. As we enter this new phase of work, we are actively seeking input from the public.  "As artificial intelligence (AI) transforms how knowledge is created, shared, and reused, we are at a fork in the road that will define the future of access to knowledge and shared creativity.  "One path leads to data extraction and the erosion of openness; the other leads to a walled-off internet guarded by paywalls.  "CC signals offer another way, grounded in the nuanced values of the commons expressed by the collective. "Based on the same prin...

"Across the country, artificial intelligence tools are changing the teaching profession as educators use them to  Help write quizzes and worksheets,  Design lessons,  Assist with grading and  Reduce paperwork.  "By freeing up their time, many say the technology has made them better at their jobs. "A poll released Wednesday by Gallup and the Walton Family Foundation found 6 in 10 U.S. teachers working in K-12 public schools used AI tools for their work over the past school year, with heavier use among high school educators and early-career teachers. "It surveyed more than 2,000 teachers nationwide in April. "Respondents who use AI tools weekly estimate they save them about six hours a week, suggesting the technology could help alleviate teacher burnout , said Gallup research consultant Andrea Malek Ash, who authored the report."

"Judge William Alsup ruled that this destructive scanning operation qualified as fair use —but only because Anthropic had legally purchased the books first, destroyed each print copy after scanning, and kept the digital files internally rather than distributing them .  "The judge compared the process to conserv[ing] space through format conversion and found it transformative.  "Had Anthropic stuck to this approach from the beginning, it might have achieved the first legally sanctioned case of AI fair use.  "Instead, the company's earlier piracy undermined its position."

"The Dream Hotel by Laila Lalami: Archivist and mother Sara T. Hussein gets detained at an airport. Her crime? A dream deemed too high risk by an AI algorithm.   "Writing incisively, Laila Lalami brilliantly builds a world where a pre-crime system collides with surveillance capitalism.  "With the novel's compelling cast of characters and endless parallels to today, I found The Dream Hotel instructive for navigating a society beset by mass surveillance —where the only escape can be found in shouldering risk together."

"They have ridiculous Italian-sounding names (like Bombombini Gusini and Trippi Troppi), and typically appear in videos on TikTok accompanied by fast-paced, AI-generated and Italian-ish (though also nonsensical) narration. "It is, in short, a meme beloved by the emerging generation Alpha (born from 2010 to 2025) and the youngest members of gen Z (generally those born from 1997 to 2012).  "And if you are any older —even if you fancy yourself as highly online , or a meme connoisseur —it is all but certain to make very little sense to you, as Tim found out at his laptop."

"We introduce AlphaGenome, a new artificial intelligence (AI) tool that more comprehensively and accurately predicts how single variants or mutations in human DNA sequences impact a wide range of biological processes regulating genes. "This was enabled, among other factors, by technical advances allowing the model to process long DNA sequences and output high-resolution predictions. "To advance scientific research, we’re making AlphaGenome available in preview via our AlphaGenome API for non-commercial research, and planning to release the model in the future. "We believe AlphaGenome can be a valuable resource for the scientific community, helping scientists better understand genome function, disease biology, and ultimately, drive new biological discoveries and the development of new treatments."

"Millions of people use ChatGPT for help with daily tasks, but for a subset of users, a chatbot can be more of a hindrance than a help. "Some people with obsessive compulsive disorder (OCD) are finding this out the hard way. "On online forums and in their therapists’ offices, they report turning to ChatGPT with the questions that obsess them, and then engaging in compulsive behavior —in this case, eliciting answers from the chatbot for hours on end —to try to resolve their anxiety."

"When I first set out to write the book, I actually didn't know it would focus so much on OpenAI.  "I really wanted to first tell the story of how the AI industry was increasingly becoming a new form of empire. "I realized as I was mapping out how I would tell this story and make it feel real and concrete and have some kind of backbone to guide readers through this epic, decades-long journey that has culminated in this frenzy that we see in the last few years, I had to do it through OpenAI. "So I changed my plan for the book and was like, okay, in addition to reporting out the history and the impacts all around the world that I want people to see to really concretize what I mean when I say empire, I also need to do a lot of insider reporting.  "To just map out what decisions were people making within this company, that then had ripple effects that shaped the AI industry and the rest of the world ."

"Hertz is introducing scanners that use artificial intelligence software to scan cars before and after they are rented to more accurately and quickly assess damage. "Hertz aims to add the equipment to 100 locations in the U.S. by the end of 2025. "A report from The Drive said that a reader who experienced the system was asked to pay $190 in processing and administrative fees to repair a roughly one-inch scrape on a wheel."

"This article sets off for an exploration of the still evolving discourse surrounding artificial intelligence (AI) in the wake of the release of ChatGPT. "It scrutinizes the pervasive narratives that are shaping the societal engagement with AI, spotlighting key themes such as agency and decision-making, autonomy, truthfulness, knowledge processing, prediction, general purpose, neutrality and objectivity, apolitical optimization, sustainability game-changer, democratization, mass unemployment, and the dualistic portrayal of AI as either a harbinger of societal utopia or dystopia.  "Those narratives are analysed critically based on insights from critical computer science, critical data and algorithm studies, from STS, data protection theory, as well as from the philosophy of mind and semiotics.  "To properly analyse the narratives presented, the article first delves into a historical and technical contextualisation of the AI discourse itself. The article then introdu...

"As thousands of applications flood job posts, 'hiring slop' is kicking off an AI arms race. "Employers are drowning in AI-generated job applications, with LinkedIn now processing 11,000 submissions per minute —a 45 percent surge from last year, according to new data reported by The New York Times . "Due to AI, the traditional hiring process has become overwhelmed with automated noise.  "It's the résumé equivalent of AI slop —call it hiring slop , perhaps —that currently haunts social media and the web with sensational pictures and misleading information.  "The flood of ChatGPT-crafted résumés and bot-submitted applications has created an arms race between job seekers and employers, with both sides deploying increasingly sophisticated AI tools in a bot-versus-bot standoff that is quickly spiraling out of control."

"AI data centers are being approved at a breakneck pace across the country, particularly in poorer regions where they are pitched as economic development projects to boost property tax receipts, bring in jobs and where they’re offered sizable tax breaks .  "Data centers typically don’t hire many people, though, with most jobs in security and janitorial work, along with temporary construction work.  "And the costs to the utility’s other customers can remain hidden because of a lack of scrutiny and the limited power of state energy regulators.  "Many data centers —like the one Meta is building in Holly Ridge —are being powered by fossil fuels.  "This has led to respiratory illness and other health risks and emitting greenhouse gasses that fuel climate change.  "In Memphis, a massive data center built to launch a chatbot for Elon Musks’ AI company is powered by smog-spewing methane turbines, in a region that leads the state for asthma rates."    

"Leading AI companies including Anthropic, Google, OpenAI and Elon Musk's xAI are discovering significant inconsistencies in how their AI reasoning models operate, according to company researchers.  "The companies have deployed chain-of-thought  techniques that ask AI models to solve problems step-by-step while showing their reasoning process, but are finding examples of misbehaviour  where chatbots provide final responses that contradict their displayed reasoning. "METR, a non-profit research group, identified an instance where Anthropic's Claude chatbot disagreed with a coding technique in its chain-of-thought but ultimately recommended it as elegant . "OpenAI research found that when models were trained to hide unwanted thoughts, they would conceal misbehaviour from users while continuing problematic actions, such as cheating on software engineering tests by accessing forbidden databases."

"Humans are different. Despite some borrowing of nomenclature from biology, neural nets used in training AI have no human-style neurons. The difference shows.   "We learn to talk and read and write with a minuscule dataset, and that process involves mimicry, emotion, cognition, and love .  "It might also have statistical weighting, but if it does, we’ve never really found that mechanism in our minds or brains. It seems unlikely that it would be there in a similar form, since these AIs have to use so much information and processing power to do what a college freshman can with a bit of motivation.  "Motivation is our problem, but it’s never a problem for AIs. They just go until their instructions reach an end point, and then they cease. AIs are unliving at the start, unliving in the process, and unliving at the end ."

"Many of us experience AI as a human we’ve built out of human metaphors.   "It’s from weirding world, a realm of spirits and oracles. We might see it as a perfect servant, happy to be subjected. Or as a friend that doesn’t judge us.  "Our metaphors are often of enchantment, bondage and servitude, it can get weird. "Sometimes we see a near-miraculous and powerful creativity, with amazing art emerging out of a machine of vectors and stats. Sometime we have the perfect slave, completely fulfilled by the opportunity to please us. Sometime we see it as an unchallenging beloved that lets us retreat from the world of real flawed humans full of feelings and flaws and blood.  "How we see it says a lot more about us than we might want to admit, but very little about AI."

"Working with big datasets lets us predict, and sometimes even ameliorate, the effects of climate on both human built systems and natural systems. "That can be anything from predicting weather, to predicting effective locations to build artificial reefs where they are most likely to revitalize ocean life. "It’s worthwhile to note that few, or maybe even none, of the powerful goods that can come from AI are consumer facing.  "None of them are the LLMs (Large Language Models) and image generators that we’ve come to know as AI.  "The benefits come from technical datasets paired with specialized AIs. Bespoke AIs can be good for a certain class of wicked problems —problems that are connected to large systems, where data is abundant and hard to understand, with dependancies (sic) that are even harder."

The U.S. National Highway Traffic Safety Administration (NHTSA) has contacted Tesla after footage shared on social media appeared to show its newly-launched driverless cars breaking traffic laws. The firm's long-awaited robotaxis —which boss Elon Musk says are central to Tesla's future —were tried out on public roads for the first time in Austin, Texas, on Sunday. Videos posted online seem to show instances where the vehicles, which had a safety driver in the passenger seat, drive erratically .

"The films shown were all short, and most would be more at home at an art film fest than something more mainstream .  "Some shorts featured an animated aesthetic (including one inspired by anime) and some presented as live action. "There was even a documentary of sorts.  "The films could be made entirely with Runway or other AI tools, or those tools could simply be a key part of a stack that also includes more traditional filmmaking methods. "Many of these shorts were quite weird. Most of us have seen by now that AI video-generation tools excel at producing surreal and distorted imagery —sometimes whether the person prompting the tool wants that or not. Several of these films leaned into that limitation, treating it as a strength."

"A new company with VC funding promises to help you fake your way through life. "Given all the cheating that’s been happening, it just makes sense that a company would come along that fully leans into AI’s primary function, which is to help our country slack off big time while pretending to work really hard.  "I guess it also makes sense that Silicon Valley’s most craven money men would think this was a great idea and proceed to throw copious amounts of cash at it. "On Friday, mega venture capital firm Andreessen Horowitz announced investment in a startup called Cluely Inc., a company whose motto is Cheat at Everything . Cluely sells what it refers to as a discrete desktop app that it claims can help people pretend (in meetings or during phone calls) like they came up with something smart to say when, in actuality, it was generated by an algorithm.   "Bloomberg reports that the VC’s investment was part of a $15 million funding round for the up-and-coming comp...

⭐AI Overview  "It's an ambitious goal to suggest that interdisciplinary investigation can eventually show how 'everything fits' in a complete and unified sense.  " Arguments for the potential of interdisciplinary investigation : Addressing complex problems : Interdisciplinary research is widely seen as necessary for tackling complex global challenges like climate change, poverty, and healthcare. Bridging knowledge gaps : It helps overcome limitations of individual disciplines by connecting insights, theories, and methods from different fields. Fostering innovation : By bringing together diverse perspectives, interdisciplinary research can spark creative solutions and lead to breakthroughs.  " Challenges and limitations : Defining 'everything' : The concept of 'everything' is vast and potentially infinite. A complete understanding of everything may remain beyond our reach. Communication barriers : Differences in terminology, methodologies, and ...

"In an ecosystem where lynxes prey on hares, a growing lynx population eventually overhunts and crashes the hare population, which then results in food scarcity and the decline of lynxes, which allows hares to recover, and so on.  "The researchers wondered if this model could also explain their fluctuating microbial ecosystems. "When they ran a version of a Lotka-Volterra model modified to introduce an outside species to the community, they found that population fluctuations made the more diverse communities more likely to be invaded.  "To be able to replicate their surprising results within a simple, time-tested model was comforting, Gore said. 'It’s telling you that you don’t need to invoke additional weird mechanisms' to explain how their microbes behaved, he said. 'It may be a surprising emergent property of these complex dynamical systems.' "However, those dynamics may not operate equally everywhere. For example, Levine, who mainly studies ...

"We introduce exacting cardinals and a strengthening of these, ultraexacting cardinals. "These are natural large cardinals defined equivalently as weak forms of rank-Berkeley cardinals, strong forms of Jónsson cardinals, or in terms of principles of structural reflection.  "However, they challenge commonly held intuition on strong axioms of infinity.  "We prove that ultraexacting cardinals are consistent with Zermelo-Fraenkel Set Theory with the Axiom of Choice (ZFC) relative to the existence of an I0 embedding.  "However, the existence of an ultraexacting cardinal below a measurable cardinal implies the consistency of ZFC with a proper class of I0 embeddings, thus challenging the linear-incremental picture of the large cardinal hierarchy.  "We show that the existence of an exacting cardinal implies that V is not equal to HOD (Gödel's universe of Hereditarily Ordinal Definable sets), showing that these cardinals surpass the current hierarchy of large c...

"For Martin Schrimpf, the promise of artificial intelligence is not in the tasks it can accomplish. It’s in what AI might reveal about human intelligence. "He is working to build a digital twin  of the brain using artificial neural networks —AI models loosely inspired by how neurons communicate with one another. "That end goal sounds almost ludicrously grand, but his approach is straightforward. First, he and his colleagues test people on tasks related to language or vision. Then they compare the observed behavior or brain activity to results from AI models built to do the same things. Finally, they use the data to fine-tune their models to create increasingly humanlike AI. "The process works best with more data and more models, so Schrimpf built an open-source platform called Brain-Score that contains nearly a hundred human neural and behavioral data sets.  "Researchers have tested thousands of AI models against the human data since Schrimpf first developed t...

Tesla's autonomous ride-hailing service is finally live. The Tesla Robotaxi launch happened on June 22, with Tesla opening the service to the public in Austin, Texas, following a little over a month of beta testing conducted by Tesla employees .  The service allows passengers to be ferried by autonomous Tesla vehicles (currently Tesla Model Y units).   The above comes with many caveats, though. The service is only operating in a small part of South Austin, and it's only open to members of the "Early Access" group, making it a sort of closed beta program.  Finally, while the cars drive autonomously, a Tesla employee must be sitting in the passenger seat, monitoring the ride.