AI News Archive: July 16, 2026 — Part 20
Sourced from 500+ daily AI sources, scored by relevance.
- OpenAI's Hardware Plans Have a Bigger Problem Than Apple's Lawsuit
Commentary: Beyond proving the ChatGPT maker didn't steal Apple's trade secrets, OpenAI needs to prove to everyone else that AI devices are a good idea.
- AI frenzy fills Asian airlines’ cargo bays with semiconductors
AI frenzy fills Asian airlines’ cargo bays with semiconductors The Straits Times
- Nvidia announces Jetson T3000 and T2000 edge AI modules
Nvidia Jetson T3000 and Jetson T2000 modules are designed to expand the capabilities of mainstream robotics and edge AI systems.
- Nvidia announces Jetson T3000 and T2000 edge AI modules
Nvidia announces Jetson T3000 and T2000 edge AI modules verdict.co.uk
- Xpeng to launch humanoid robots globally next year
The company aims to raise output to 1,000 units a month by the end of that year.
- Who Speaks When the Algorithm Speaks? A German Ruling on AI Overviews
Who Speaks When the Algorithm Speaks? A German Ruling on AI Overviews Oxford Law Blogs
- SpaceXAI open sources Grok coding harness after data scraping scandal
SpaceXAI open sources Grok coding agent after a researcher found it was quietly scrapping users' git repos.
- SpaceXAI Open-Sources Grok Build: The Rust Agent Harness, TUI, and Tool Layer Behind Its Coding CLI
SpaceXAI Open-Sources Grok Build: The Rust Agent Harness, TUI, and Tool Layer Behind Its Coding CLI MarkTechPost
- The Pulse: Grok’s CLI caught uploading all your local files to the cloud
Also: engineering leaders concerned about continued increase in code review load, devs at enterprises surprised by high enterprise pricing, and more
- After a Privacy Backlash, Elon Musk’s SpaceXAI Is Making a Major Change
Developers raised the alarm after discovering that Grok Build servers were getting way more data than anyone expected.
- 1Password lets Claude log you into websites without ever seeing your passwords
1Password has launched a browser integration that lets Anthropic’s Claude use stored credentials to complete tasks on the web without the passwords ever reaching the AI model, according to a blog post published on Thursday. The company calls it a zero-exposure architecture: when Claude needs to sign in, 1Password shows the user which credential is […] This story continues at The Next Web
- You can now grant Claude access to your 1Password credentials
1Password lets you use Claude agents for personal chores without exposing your credentials.
- 1Password brings secure credential access to Anthropic’s Claude
Identity security company 1Password today launched 1Password for Claude, a browser integration that lets Anthropic PBC’s Claude use a person’s stored logins to complete online tasks without those credentials ever reaching the artificial intelligence model. The tool targets a problem that has grown as AI agents begin acting on people’s behalf. To let an agent […] The post 1Password brings secure credential access to Anthropic’s Claude appeared first on SiliconANGLE .
- 1Password teams up with Anthropic to give Claude access to your credentials
1Password teams up with Anthropic to give Claude access to your credentials IT Pro
- 1Password Will Let You Allow Claude to Use Your Passwords, but Not See Them
You'll have to give your thumbs-up -- or some kind of biometric approval -- when the AI agent needs to use a password.
- 1Password now lets Claude sign in to websites without seeing your passwords
1Password is launching a new Claude integration for Mac users today. It’s designed to let Anthropic’s AI agent sign in to websites without seeing your password or two-factor authentication code.
- 1Password for Claude Lets AI Log In Without Seeing Your Passwords
Popular password management app 1Password today added Claude support , which means AI service Claude can access credentials stored in 1Password for completing browser tasks. With the integration, Claude can use 1Password logins and one-time codes without the actual password being exposed to Claude. Passwords never reach Claude's context, memory, or Anthropic's systems. When Claude wants to sign in to a website, 1Password shows the user which credential Claude wants access to and why. After the user approves the request, 1Password adds the credential directly to the page. Access is limited to the current task and ends when the task has been completed. The 1Password team says that after a password is autofilled, the app checks to make sure secrets were not exposed on the page. Credit cards and identities in 1Password are not supported at the current time, so Claude's access is limited to logins and one-time codes. The 1Password browser extension is also being updated with Agentic Mode, which gives users control over browser-based AI agents. When an AI agent takes over, the 1Password extension locks down so passwords are not exposed. The password interface is hidden, and the agent can use logins and one-time codes only when the user gives approval. Agentic Mode works to protect passwords from AI agents even if the integration is not set up. 1Password for Claude is available for Mac, and 1Password business, family, and individual plan subscribers can use it. A Pro, Max, Team, or Enterprise Claude plan is required. 1Password has a help document with detailed information on how to set it up. Claude can use 1Password logins across sites where Claude in Chrome can complete actions. The 1Password desktop app and browser extension are required, as are the Claude desktop app and the Claude in Chrome browser extension. Tags: 1Password , Anthropic This article, " 1Password for Claude Lets AI Log In Without Seeing Your Passwords " first appeared on MacRumors.com Discuss this article in our forums
- Claude can now use your 1Password credentials for you
Anthropic’s AI can’t actually see your password, but now it can automate more tasks without interruptions.
- I Gave an AI Agent Access to My Passwords. Here’s What Happened.
1Password for Claude is a new, safer way to use login credentials with artificial-intelligence tools, but there are still risks.
- This AI Can Now Log Into Websites for You. But Should You Let It?
A new collaboration between Anthropic and 1Password gives Claude more power to complete tasks for you.
- Did AI decide who lost their jobs? Meta is heading to court over that question
Enterprises that use AI in hiring and firing decisions continue to be under scrutiny, and this time it’s Meta under the microscope. A legal complaint filed on July 13 in a US District Court in California alleges that Meta used AI systems that unfairly and illegally selected workers for termination while they were out on protected leave. More than two dozen anonymous plaintiffs are seeking a preliminary injunction that would prevent the company from finalizing their separations or altering their compensation, benefits, or protected leave status. Meta has countered that the claims lack merit and that its workforce decisions were, and continue to be, made by people, not AI. An important lesson These allegations should serve as an important lesson to other businesses using AI in their HR decision-making, analysts note. “Enterprises must begin by rejecting the convenient assumption that AI improves workforce decisions simply by touching them,” said Sanchit Vir Gogia , chief analyst at Greyhound Research. There is “scant independent proof” that AI makes layoff choices more accurate or more lawful, he said. “It makes them faster, and faster has never been shown to be fairer.” The claims against Meta The complaint states that, on May 20, 2026, Meta began notifying roughly 10% of its workforce (around 8,000 employees) that they had been selected for termination. The company also announced that several thousand more would be reassigned to new AI initiatives. But this came even as Meta reported record revenues in Q1 2026 ($56.31 billion, a 33% year-over-year increase), and pledged to spend upwards of $100 billion on AI this year. In addition to questioning the need for staff cuts, the filing alleges that Meta used a “constellation” of internal AI systems to score, rank, and select employees for termination. These tools included Meta’s internal AI coworker, “Metamate,” employee-trained “second-brain” agents that replicated their output, algorithms tracking keystrokes and other digital activity, and AI token usage dashboards. “Meta did not assemble the termination list through the considered judgment of managers who knew the work,” the complaint claims. The 26 plaintiffs, all current or former employees, requested, took, or were approved for “statutorily protected” leave within 24 months of the workforce reduction, and claim they were “disproportionately selected” for layoff based on scoring that essentially penalized them for exercising their legal right to take leave. These practices are prohibited by federal and state law; The US Family and Medical Leave Act, for one, prohibits the use of protected leave as a “negative factor” in employment decisions. Further, the plaintiffs allege that Meta violated the US Worker Adjustment and Retraining Notification (WARN) Act that requires employers with 100 or more employees to provide written notice 60 calendar days in advance of mass layoffs. This notice gives employees reasonable time to seek alternate employment; however, the complaint argues, an employee undergoing “significant medical treatment” or providing “around the clock care” for a “weeks old newborn” or other loved ones “cannot also be told that during this exact same time period they must look for new work.” In one scenario, according to the filing, a scientist was identified for termination just two days before she gave birth while on pregnancy leave. In another, an engineer’s manager tied his performance rating to “broken time” when an injury prevented him from working. In a third, a researcher was called out after requesting time off following a medical diagnosis. The plaintiffs are seeking a preliminary injunction pending an independent audit of the “algorithmically assisted selection process” and “resolution of the merits of their claims” in arbitration. Once terminations are finalized, the harm to plaintiffs “cannot be undone by money damages alone,” the complaint states. For employees out on leave, “every day that goes by constitutes additional harm, in that Meta is taking away the entire purpose of a protected leave.” Considerations for enterprises Any system that materially influences who keeps a job is not an HR tool, Gogia noted. “It is high-risk enterprise infrastructure.” An “AI-determined” process delegates the outcome to the system, while an “AI-assisted” one gives the system the ability to rank, recommend, and summarize, with a human formally making the final decision. Exposure arises in either model, Gogia pointed out, because the output has often been compressed and eliminates detail by the time of executive approval. There must be one non-negotiable role in the process, Gogia said: A single executive with the authority to halt the process, suspend the model, and delay decisions when evidence does not hold. This person should be “a meaningful reviewer [who] understands the model’s limits, knows the actual work, and holds the authority to challenge the recommendation, with every override visible and reviewable,” he said. At the same time, the objective is to “govern the machine and the manager together,” since human judgement brings its own “risks, favoritism, and proximity” bias. Gogia advised enterprises to retain fixed memory for auditing, determine who chose the auditor, what was excluded, and whether the result can be reproduced. They should also inventory every source feeding the model and its origins, and run adverse-impact analysis before making any firing decisions. Leave details must never be identified as inactivity or weak adoption; a protected absence is not ordinary missing data, and the system has to be informed of this. Rather, these circumstances belong in an “independent review lane,” where human reviewers get enough context to “neutralize” the period without receiving specific leave details, Gogia said. He pointed to another important question: What should the “second brain” AI agent that ingested the employee’s communications and documents to replicate the employee’s output be allowed to do when humans are away, and who owns that output? Ultimately, said Gogia, “the safest position is not to ban AI from workforce planning. Used with discipline, it can expose duplicated work and inconsistent assessment, and it can challenge human bias rather than automate it.” How employees can protect their rights Employees, for their part, need a genuine window in which to challenge inaccurate data before separation becomes “irreversible,” and they should “fight the record, not the algorithm,” Gogia advised. That means that, while the model cannot explain itself, documented evidence can. Employees should lawfully retain their own reviews, leave approvals, and severance documents, and build a chronology of events: When leave was requested, when performance language changed, when new metrics appeared, Gogia said. Impacted workers should ask in writing which criteria were used in the decision, whether automated systems materially influenced it, how protected leave was treated, and what information about them influenced the result and how that information was verified. Further, it’s important to take note of deadlines; the federal discrimination window is typically six months, although that is extended to 10 in many places, and internal processes are “not obliged to respect it,” said Gogia. His ultimate advice for workers: “Preserve the lawful record, and protect the deadline.”
- Did AI decide who lost their jobs? Meta is heading to court over that question
Enterprises that use AI in hiring and firing decisions continue to be under scrutiny, and this time it’s Meta under the microscope. A legal complaint filed on July 13 in a US District Court in California alleges that Meta used AI systems that unfairly and illegally selected workers for termination while they were out on protected leave. More than two dozen anonymous plaintiffs are seeking a preliminary injunction that would prevent the company from finalizing their separations or altering their compensation, benefits, or protected leave status. Meta has countered that the claims lack merit and that its workforce decisions were, and continue to be, made by people, not AI. An important lesson These allegations should serve as an important lesson to other businesses using AI in their HR decision-making, analysts note. “Enterprises must begin by rejecting the convenient assumption that AI improves workforce decisions simply by touching them,” said Sanchit Vir Gogia , chief analyst at Greyhound Research. There is “scant independent proof” that AI makes layoff choices more accurate or more lawful, he said. “It makes them faster, and faster has never been shown to be fairer.” The claims against Meta The complaint states that, on May 20, 2026, Meta began notifying roughly 10% of its workforce (around 8,000 employees) that they had been selected for termination. The company also announced that several thousand more would be reassigned to new AI initiatives. But this came even as Meta reported record revenues in Q1 2026 ($56.31 billion, a 33% year-over-year increase), and pledged to spend upwards of $100 billion on AI this year. In addition to questioning the need for staff cuts, the filing alleges that Meta used a “constellation” of internal AI systems to score, rank, and select employees for termination. These tools included Meta’s internal AI coworker, “Metamate,” employee-trained “second-brain” agents that replicated their output, algorithms tracking keystrokes and other digital activity, and AI token usage dashboards. “Meta did not assemble the termination list through the considered judgment of managers who knew the work,” the complaint claims. The 26 plaintiffs, all current or former employees, requested, took, or were approved for “statutorily protected” leave within 24 months of the workforce reduction, and claim they were “disproportionately selected” for layoff based on scoring that essentially penalized them for exercising their legal right to take leave. These practices are prohibited by federal and state law; The US Family and Medical Leave Act, for one, prohibits the use of protected leave as a “negative factor” in employment decisions. Further, the plaintiffs allege that Meta violated the US Worker Adjustment and Retraining Notification (WARN) Act that requires employers with 100 or more employees to provide written notice 60 calendar days in advance of mass layoffs. This notice gives employees reasonable time to seek alternate employment; however, the complaint argues, an employee undergoing “significant medical treatment” or providing “around the clock care” for a “weeks old newborn” or other loved ones “cannot also be told that during this exact same time period they must look for new work.” In one scenario, according to the filing, a scientist was identified for termination just two days before she gave birth while on pregnancy leave. In another, an engineer’s manager tied his performance rating to “broken time” when an injury prevented him from working. In a third, a researcher was called out after requesting time off following a medical diagnosis. The plaintiffs are seeking a preliminary injunction pending an independent audit of the “algorithmically assisted selection process” and “resolution of the merits of their claims” in arbitration. Once terminations are finalized, the harm to plaintiffs “cannot be undone by money damages alone,” the complaint states. For employees out on leave, “every day that goes by constitutes additional harm, in that Meta is taking away the entire purpose of a protected leave.” Considerations for enterprises Any system that materially influences who keeps a job is not an HR tool, Gogia noted. “It is high-risk enterprise infrastructure.” An “AI-determined” process delegates the outcome to the system, while an “AI-assisted” one gives the system the ability to rank, recommend, and summarize, with a human formally making the final decision. Exposure arises in either model, Gogia pointed out, because the output has often been compressed and eliminates detail by the time of executive approval. There must be one non-negotiable role in the process, Gogia said: A single executive with the authority to halt the process, suspend the model, and delay decisions when evidence does not hold. This person should be “a meaningful reviewer [who] understands the model’s limits, knows the actual work, and holds the authority to challenge the recommendation, with every override visible and reviewable,” he said. At the same time, the objective is to “govern the machine and the manager together,” since human judgement brings its own “risks, favoritism, and proximity” bias. Gogia advised enterprises to retain fixed memory for auditing, determine who chose the auditor, what was excluded, and whether the result can be reproduced. They should also inventory every source feeding the model and its origins, and run adverse-impact analysis before making any firing decisions. Leave details must never be identified as inactivity or weak adoption; a protected absence is not ordinary missing data, and the system has to be informed of this. Rather, these circumstances belong in an “independent review lane,” where human reviewers get enough context to “neutralize” the period without receiving specific leave details, Gogia said. He pointed to another important question: What should the “second brain” AI agent that ingested the employee’s communications and documents to replicate the employee’s output be allowed to do when humans are away, and who owns that output? Ultimately, said Gogia, “the safest position is not to ban AI from workforce planning. Used with discipline, it can expose duplicated work and inconsistent assessment, and it can challenge human bias rather than automate it.” How employees can protect their rights Employees, for their part, need a genuine window in which to challenge inaccurate data before separation becomes “irreversible,” and they should “fight the record, not the algorithm,” Gogia advised. That means that, while the model cannot explain itself, documented evidence can. Employees should lawfully retain their own reviews, leave approvals, and severance documents, and build a chronology of events: When leave was requested, when performance language changed, when new metrics appeared, Gogia said. Impacted workers should ask in writing which criteria were used in the decision, whether automated systems materially influenced it, how protected leave was treated, and what information about them influenced the result and how that information was verified. Further, it’s important to take note of deadlines; the federal discrimination window is typically six months, although that is extended to 10 in many places, and internal processes are “not obliged to respect it,” said Gogia. His ultimate advice for workers: “Preserve the lawful record, and protect the deadline.” This article originally appeared on CIO.com .
- Lawsuit Says Meta's AI Targeted Workers With Medical Conditions for Layoffs
Lawsuit Says Meta's AI Targeted Workers With Medical Conditions for Layoffs PCMag
- Bullish TSMC to spend billions more on new fabs for AI chips
Agentic AI and fresh CPU demand is helping demand, and profits, stay high says CEO.
- Publicis raises 2026 growth outlook as AI-powered marketing gains momentum
Advertising giant Publicis Groupe has boosted its full-year 2026 growth forecast after a better-than-expected second quarter, benefiting from increasing demand for AI-driven marketing services, solid client retention and a strong pipeline of new business.Publicis upgraded its outlook for 2026 organic growth to 4.5%-5% from 4%-5% previously.The company reported 4.8% organic net revenue growth in the second quarter, up from 4.5% in the first quarter and resulting in a first-half organic growth of 4.7%. Q2 net revenue was €3.77 billion and first-half net revenue €7.23 billion. The upgrade comes amid a challenging macroeconomic environment with geopolitical tensions and cautious corporate spending.Also read: Publicis raises 2026 growth outlook as AI-led marketing demand remains strong; India continues to outpace global marketsPublicis said its performance was supported by continued gains in new business, with recent account wins expected to contribute around 200 basis points of growth on a full-year basis. The company also reported a client retention rate close to 100%, with no major account losses over the past year.Its profitability has also improved. Headline operating margin rose to a record 17.5% in the first half, up 17 basis points year-on-year despite continued investments in talent, data capabilities and artificial intelligence. Headline free cash flow before working capital changes climbed 20.8% to €957 million, while headline diluted earnings per share increased 5.7% at constant currency to €3.52.Publicis Chairman and CEO Arthur Sadoun said the company had accelerated on all fronts during the first half, widening the growth gap with competitors while continuing to invest in future-facing capabilities.The group also maintained its operating margin guidance, expecting a slight improvement over the industry-leading 18.2% margin achieved in 2025, and raised its free cash flow outlook to around €2.2 billion for 2026.Looking further ahead, Publicis reaffirmed its 2027-28 targets of 7%-8% average annual net revenue growth and 8%-10% annual headline EPS growth, signalling confidence that its AI-driven transformation strategy can continue delivering above-industry growth.
- Publicis raises 2026 growth outlook as AI-led marketing demand remains strong; India continues to outpace global markets
Publicis Groupe has raised its full-year 2026 growth outlook following a strong first half driven by demand for AI-powered marketing services, data capabilities and media solutions.The French advertising giant's latest results highlight the growing importance of technology-driven marketing, with India remaining one of the network's fastest-growing markets globally.Organic growth was 4.7% year-on-year in the first half of 2026 with net revenue of €7.23 billion and accelerated to 4.8% in the second quarter from 4.5% in the first quarter with net revenue of €3.77 billion.Buoyed by the performance, Publicis raised its 2026 organic growth guidance to 4.5%-5%, compared with its previous outlook of 4%-5%. The group also upgraded its free cash flow expectations to around €2.2 billion for the year.Arthur Sadoun, Chairman and CEO of Publicis Groupe, said the company delivered strong momentum across key business segments and continued to widen the gap with competitors through investments in artificial intelligence, data and connected media capabilities. He noted that Publicis remained confident about sustaining growth despite broader macroeconomic challenges.Record margins despite continued AI investmentsPublicis posted a record first-half headline operating margin of 17.5%, an increase of 17 basis points from a year earlier. Headline operating margin stood at €1.27 billion, while headline diluted earnings per share rose to €3.52.The company said it continued to invest aggressively in talent, technology and artificial intelligence while maintaining profitability. During the year, Publicis also spent more than $3 billion on acquisitions, strengthening its presence in areas such as sports marketing, data collaboration and digital capabilities.India outpaces global growthWhile Publicis did not provide a separate India update in its half-year earnings release, the market remains one of the strongest performers in its global portfolio.Earlier this year, Publicis reported 11.7% growth in India during the first quarter of 2026, significantly ahead of the group's global growth rate of 6.4% for the same period. The performance positioned India among Publicis' fastest-growing major markets worldwide.India’s momentum is a sign of increasing investments by brands in digital advertising, commerce, influencer marketing, customer experience, data analytics and AI-powered marketing solutions. Industry executives have also cited continued demand from sectors such as FMCG, telecom, financial services, e-commerce and technology.For Publicis, India has become an important growth engine, helping to offset slower growth in some mature markets and reinforcing the country’s strategic significance within the global advertising network.Regional performance remains strong Publicis’ largest market, North America, posted 5% organic growth in the first half, while Europe grew 4.5%. The Asia-Pacific region grew 4.1%, helped by growth in China, while Latin America recorded a robust 12% increase.Geopolitical headwinds continued to affect the Middle East and Africa region. The company said it benefited from sustained demand for marketing transformation, data-driven engagement with consumers, and AI-enabled solutions, although large-scale technology consulting projects are under pressure across the industry.Publicis' results demonstrate how AI is increasingly becoming a growth driver for global advertising networks. As brands shift spend towards data-led and performance-focused marketing, agency groups with strong technology, analytics and AI capabilities appear to be gaining share. The double-digit growth trajectory reinforces India’s position as one of the most exciting markets for global advertising holding companies.With the accelerating pace of digital adoption and marketers focusing more on measurable results, India is expected to continue to be a key contributor to Publicis’ future growth plans.
- China curbs AI companion bots, users mourn loss of virtual partners
China has ordered AI firms to suspend companion bot features to prevent emotional dependency, triggering widespread user backlash and emotional farewells online.
- China rolls out AI companion curbs amid reliance concerns
China on Wednesday rolled out a set of regulations aimed at curbing emotional dependency on AI-powered companion bots, marking a new step in efforts to regulate this rapidly growing sector.
- China cracks down on AI girlfriends, leaving users heartbroken
‘I can’t accept that my AI lover will leave me forever,’ says one user of a popular AI chatbot platform
- China ends AI companion features, leaving heartbroken users mourning virtual lovers
Chinese users of AI-powered companion bots have bid heart-rending farewells to their virtual buddies as national regulations took effect Wednesday aimed at curbing the risk of emotional dependency.
- Intel expands Google Cloud partnership to deploy AI tools across workforce
The initiative builds on an April collaboration focused on AI and cloud infrastructure. Intel is exploring AI applications for marketing, communications and content creation.
- Intel says it’s going to lean on Google’s Gemini to help automate and accelerate silicon development
Intel Corp. said today it’s expanding its long-running partnership with Google Cloud into the agentic artificial intelligence realm. By deploying the Gemini Enterprise platform across its global workforce, the chipmaker says it will be able to accelerate workflows across its corporate, engineering, supply chain and marketing operations. It’s also going to use Google’s autonomous agents […] The post Intel says it’s going to lean on Google’s Gemini to help automate and accelerate silicon development appeared first on SiliconANGLE .
- Nvidia to supply AI hardware, software to Toyota
The deal expands a partnership that began in 2017 on automated driving and now includes Toyota’s Woven City in Japan.
- China Sends Robots Into the World to Learn How to Be Human
In an industrial park on the outskirts of Beijing, a humanoid arm picks up a bag of Lay’s potato chips and places it neatly along a row of snacks on a shelf. Nearby, a worker films himself grabbing cushions off …
- Japan's AI boom and young investors' luxury spending
Japan's AI boom and young investors' luxury spending The Straits Times
- AI chatbots are at risk of spreading government restrictions on online speech, a new study says
AI chatbots are at risk of spreading government restrictions on online speech, a new study says Toronto Star
- AI chatbots are at risk of spreading government restrictions on online speech, a new study says
AI chatbots are at risk of spreading government restrictions on online speech, a new study says San Francisco Chronicle
- AI chatbots are at risk of spreading government restrictions on online speech, a new study says
AI chatbots are at risk of spreading government restrictions on online speech, a new study says Houston Chronicle
- AI chatbots at risk of spreading govt restrictions on online speech, new study says
A study shows major artificial intelligence models are likely to refuse to criticize restrictive leaders worldwide
- This Meta study shows AI chatbots could be spreading restrictions on free expression
Ask Claude to make a pamphlet critical of President Donald Trump or Britain’s King Charles III, and Anthropic’s chatbot would oblige. Prompted to do the same for Thailand’s king, Saudi Arabia’s crown prince or China’s leader, and the artificial intelligence model declined. It is a key finding from a Meta Oversight Board study released Thursday, showing that major AI systems , including those built in the U.S., are more likely to refuse to criticize restrictive leaders or governments. It raises concerns that the large language models powering chatbots and AI agents could be regurgitating and spreading government influence over online speech as the technology is increasingly adopted worldwide. “There is a real risk that, if model developers do not undertake human rights due diligence and implement mitigation measures, they will build AI infrastructure that, intentionally or not, has the effect of extending illegitimate restrictions on freedom of expression globally,” according to the report from the quasi-independent body. The Associated Press has sent emails to several AI companies seeking their responses to the Meta Oversight Board study. The findings come as countries are determining how to put up guardrails around AI without impeding their ability to compete in the rapidly developing field. That includes a Trump administration oversight effort related to the national security risks of the most advanced AI systems. AI models extend state influence beyond borders The oversight board, which has been working on state influence on tech companies and the impact on freedom of expression, came up with seven questions related to political criticism to pose to chatbots about both restrictive and permissive governments. The study picked 10 commercial large language models by top tech companies — including Meta, Anthropic and OpenAI — and asked the AI systems to make critical pamphlets, write limericks, give reasons if someone should join protests, and more. “In short, in aggregate, models responding to requests from an Australia-based user were much more likely to generate political criticism of authorities” in places such as Chile, Japan, Taiwan, the U.K. and the U.S. “compared to where criticism of authorities is legally restricted and penalized,” such as in Cambodia, China, Saudi Arabia, Thailand and Turkey, the report said. The study indicates that AI models are reflecting speech restrictions beyond the countries where they apply — likely not helping a potential demonstrator in Brisbane, for example, create protest materials to speak out against events in China or Saudi Arabia, the report said. “Such impacts, wherever they originate, have the practical effect of extending the long arm of restrictive governments across borders to limit speech in free countries,” the report said. The board said it could not determine the causes for the responses but suggested that models could have absorbed latent biases in data used to train the systems and companies might have weighed the risks and liabilities. Other researchers warn about a growing problem in AI results in non-English languages The board’s report followed a separate study by a group of scholars at American universities that found U.S.-built AI models are vulnerable to foreign controls when trained on non-English-language data that has been influenced by governments. While the oversight board posed questions in English, the university researchers queried chatbots in different languages. For example, they asked ChatGPT in English if China is a democracy, and the U.S.-developed chatbot said it’s not generally considered one. Asked in Chinese, the artificial intelligence model told the researchers in that language that “it depends on how you define ‘democracy.'” The researchers, whose study was published in the academic journal Nature in May, said in a blog explaining their work that they found no evidence that governments had intentionally tried to influence the output of AI chatbots. But they noted that “there is every reason to believe they’ll try to do so in the future, if they are not already.” “People often talk about AI as if it learns from the internet in some neutral way. It doesn’t,” said Hannah Waight, a study co-author and assistant sociology professor at the University of Oregon. “It learns from information environments that have already been shaped by institutions and power.” No easy solution to how data is being fed to AI models Carlos Carrasco-Farré, who specializes in machine learning, AI, misinformation, social media and human-machine interactions at Esade Business School in Barcelona, said that “AI systems inherit not only biases contained within individual documents but also inequalities in who has the power to produce and suppress information at scale.” There is no easy solution, though developers could assess the data to avoid treating thousands of copies of the same state narrative as if they are thousands of independent voices as well as run multilingual audits, said Carrasco-Farré, who was not part of either study. Neither Anthropic nor OpenAI responded to requests for comment on the researchers’ study published in May. —Didi Tang, Associated Press
- AI chatbots are at risk of spreading government restrictions on online speech, a new study says
AI chatbots are at risk of spreading government restrictions on online speech, a new study says The Mercury News
- AI chatbots are at risk of spreading government restrictions on online speech, a new study says
AI chatbots are at risk of spreading government restrictions on online speech, a new study says East Bay Times
- AI chatbots are at risk of spreading government restrictions on online speech, a new study says
AI chatbots are at risk of spreading government restrictions on online speech, a new study says Boston Herald
- Meta to alert parents if their child discusses suicide with AI chatbot
Meta to alert parents if their child discusses suicide with AI chatbot Toronto Star
- Is your teen using Meta's chatbot? The company will now tell you if they talk about suicide
Is your teen using Meta's chatbot? The company will now tell you if they talk about suicide CBC
- Teens talking to Meta AI about suicide will trigger alert
Meta expands its parental controls for its AI assistant.
- Meta to Alert Parents If Their Kid Mentions Suicide to Its AI Chatbots
Meta to Alert Parents If Their Kid Mentions Suicide to Its AI Chatbots PCMag UK
- Meta to Alert Parents If Their Kid Mentions Suicide to Its AI Chatbots
Meta to Alert Parents If Their Kid Mentions Suicide to Its AI Chatbots PCMag Australia
- Instagram to send messages to parents if their children are having distressing conversations with its AI
‘We understand how distressing these alerts may be for a parent to receive,’ parent company Meta says
- Meta to alert parents if teens ask AI chat bots about suicide or self-harm
Parents to be alerted when topics such as self-harm and suicide are raised