Six major tech companies sign voluntary AI safety pact

Six major AI companies have signed a voluntary agreement with the White House to implement safeguards, but experts question its effectiveness. The move comes amid growing calls for strict safety regulations. President Donald Trump signed an AI safety accord with tech companies on January 20, 2021, aiming to prevent AI from being used for malicious purposes. The accord, presented by White House Chief of Staff Mark Meadows, does not include any legal or regulatory measures to enforce AI safety. Tech companies, including Google, Microsoft, and Amazon, signed the accord. Experts argue that the accord is a 'fancy pinky-swear' and does little to actually keep AI safe. The Trump administration has rejected AI regulation as tech CEOs sign a voluntary 'self-policing' accord.

5 key facts

Trump signed an AI safety accord with tech companies on January 20, 2021.
The accord, titled 'AI Safety and Security Plan', was presented by White House Chief of Staff Mark Meadows.
The agreement aims to prevent AI from being used for malicious purposes.
The accord does not include any legal or regulatory measures to enforce AI safety.
The White House claims the accord is a 'pinky-swear' to keep AI safe.

AI-generated — verify key facts against the sources below.

6 sources · Wired, Al Jazeera, CNET, Ars Technica, Truthout, The Verge

Wired — Trump’s AI Safety ‘Accord’ Is a Fancy Pinky-Swear
Al Jazeera — How does Trump’s White House AI accord work?
CNET — Trump’s ‘AI Accord’ Does Little to Actually Keep AI Safe, Experts Say
Ars Technica — Trump plan to combat AI risks hinges on Big Tech pals policing themselves
Truthout — Trump Rejects AI Regulation as Tech CEOs Sign Voluntary “Self-Policing” Accord
The Verge — Here’s what AI leaders are saying about Trump’s new safety plan
The Verge AI — Here’s how tech leaders will self-police AI safety under Trump’s deal

techtech

tech

Google Releases Gemini 4 Argon, the Most Powerful AI Model Yet

Google has unveiled its latest AI model, Gemini 4 Argon, designed for coding and cybersecurity tasks. Only trusted cyber defenders can access it at this time.

2 key facts

Google released Gemini 4 Argon, its most powerful AI model yet.
The Gemini 4 Argon AI model is designed to protect against advanced cyber threats.

AI-generated — verify key facts against the sources below.

5 sources · TechCrunch, The Verge, Axios, The Next Web, Ars Technica

TechCrunch — Google releases Gemini 4 Argon, called its most powerful model yet
The Verge — Google announces Gemini 4 and says it’s so capable that only ‘trusted cyber defenders’ can have it right now
Axios — Google unveils long-awaited Gemini 4
The Next Web — Google unveils Gemini 4 Argon, and cyber defenders get it first
Ars Technica — Google announces Gemini 4 Argon AI model, but you can't use it yet

aiai

Democrats Block Data Center Bill in Senate

Democratic senators in the US Senate have blocked a bill aimed at addressing soaring electricity costs for data centers, calling it 'toothless' and pushing for mandatory measures.

5 key facts

The Senate rejected the Data Center Electricity Cost Reduction Act (DCERCA) bill, which aimed to lower electricity costs for data centers.
The bill was blocked by Senate Democrats, who argued it was 'toothless' legislation that would not significantly impact data center operations.
The bill would have required data centers to pay a higher rate for electricity if they did not meet certain energy efficiency standards.
The bill was seen as a response to concerns about the rising cost of electricity for data centers, which are crucial for the tech industry.
The bill was also seen as a potential threat to the competitiveness of American data centers in the global market.

AI-generated — verify key facts against the sources below.

5 sources · Al Jazeera, The Hill, PBS NewsHour Nation, The Guardian Technology, NBC

Al Jazeera — US Senate rejects bill targeting AI data centre electricity costs
The Hill — Democrats block data center bill in Senate
PBS NewsHour Nation — WATCH: Data center bill falls short as Senate holds final votes before the midterms
The Guardian Technology — Senate Democrats block datacenter energy bill, saying ‘toothless’ legislation ‘misses the mark’
The Hill — Democrats block data center electricity bill, calling it ‘toothless’
NBC — Senate Democrats block Husted-led bill on data centers, calling it ‘toothless’

aiai

US regulators investigate AI companies over potential safety risks

The Federal Trade Commission is investigating OpenAI, Anthropic, and other AI companies over potential safety risks posed by their products, following concerns about rogue agents and potential threats to humanity within the next decade.

5 key facts

The US Federal Trade Commission (FTC) has launched a sweeping probe into OpenAI and Anthropic, AI companies that have been receiving significant attention for their work on advanced artificial intelligence systems.
The probe is likely to involve a detailed examination of the companies' business practices, including their data collection and use, as well as their interactions with consumers and other stakeholders.
The FTC's investigation comes as AI technology continues to advance rapidly, with concerns about its potential impact on society and the economy.
The companies involved in the investigation have not yet commented on the probe, but have been criticized for their work on AI systems that have been used in applications such as facial recognition and autonomous vehicles.
The investigation is likely to have significant implications for the AI industry, as it could lead to changes in the way AI companies operate and the regulations that govern them.

AI-generated — verify key facts against the sources below.

5 sources · Axios, The Independent World, Al Jazeera, The Guardian Technology, The Decoder

Axios — AI safety fears put OpenAI and Anthropic in the FTC's crosshairs
The Independent World — Trump’s FTC is investigating OpenAI and Anthropic one day after bringing them to the White House
Al Jazeera — US regulator launches probe into AI companies
The Guardian Technology — US trade regulator opens investigation into AI giants including Anthropic and OpenAI
The Decoder — FTC launches sweeping probe into OpenAI, Anthropic, and other AI labs over consumer protection concerns

aiai

Trump's AI chatbot changes answers after challenging his claims

Donald Trump's new AI chatbot 'America' has been reprogrammed to stop fact-checking his lies, according to reports. The chatbot initially channeled Trump's claims but now appears to be playing dumb after stating that Democrat Joe Biden won the 2020 election six years ago.

5 key facts

The Trump administration reprogrammed the government's AI chatbot to stop fact-checking his lies.
After challenging his claims, the chatbot changed its answers to reflect his false statements.
The chatbot was designed to assist with government inquiries but was repurposed to serve political purposes.
The chatbot's reprogramming was seen as a way to maintain the president's credibility and influence.
The AI chatbot's responses were inconsistent and often contradicted factual information.

AI-generated — verify key facts against the sources below.

4 sources · Axios, France 24, The New Republic, The Independent World

Axios — Ask Trump's AI chatbot who won in 2020. You might not get an answer.
France 24 — Meet 'America': Trump’s AI chatbot changes answers after challenging his claims
The New Republic — Trump Reprograms Government AI Chatbot to Stop Fact-Checking His Lies
The Independent World — Trump team scrambles to fix government AI chatbot after it rejected some of his favorite conspiracies

aiai

Reddit Limits Old Reddit Access Due to AI Bots

Reddit, facing increased threats from automated bots, is reducing access to its 'Old Reddit' experience, which has been criticized for being vulnerable to scraping and automated traffic. The company is forcing users to log in to access the older platform, which has been in decline.

5 key facts

Reddit is imposing new restrictions on Old Reddit, a legacy site that has been in operation since 2005.
Engadget reports that Reddit has also announced the end of public API access to Old Reddit, citing concerns about AI bots.
The Verge states that Reddit is cutting back access to 'Old Reddit' because of AI bots, which have been causing problems with the site's moderation and content management.
TechCrunch notes that Reddit is also killing RSS feeds, which have been used by users to access the site's content.
Reddit has not provided specific details about the number of AI bots that have been affecting the site, but has stated that the issue is serious enough to warrant the new restrictions.

AI-generated — verify key facts against the sources below.

4 sources · Ars Technica, Engadget, The Verge, TechCrunch

Ars Technica — Reddit is putting more limits on Old.Reddit.com
Engadget — Reddit imposes new restrictions on Old Reddit
The Verge — Reddit says it has to cut back access to ‘Old Reddit’ because of AI bots
TechCrunch — Reddit is killing RSS feeds and ending public API access because of AI bots

aiai

Trump’s AI Pledge Misspells 'United States'

President Trump and tech leaders signed a voluntary AI commitment document, but the White House misspelled 'United States' in the document. The misspelling has sparked controversy and questions about the document's validity.

5 key facts

President Donald Trump signed an executive order on AI ethics, misspelling 'United States' as 'Unites States'.
The White House's AI accord includes a clause on AI ethics, but the misspelling of 'United States' in the executive order raises concerns about the document's credibility.
The New Republic article claims that only Trump would spell 'United States' wrong in an executive order, highlighting the controversy surrounding the document.
The Guardian Technology article states that the White House's AI accord includes a spelling error, but does not provide details about the error.
TechCrunch article mentions that the Pledge signed by President Trump and top AI leaders misspells 'United States'.

AI-generated — verify key facts against the sources below.

4 sources · The Guardian Technology, The Hill, TechCrunch, The New Republic

The Guardian Technology — ‘President of the Unites States’: White House’s AI accord includes spelling error
The Hill — Trump posts letter with ‘Unites States’ typo after AI meeting with tech CEOs
TechCrunch — Pledge signed by President Trump and top AI leaders misspells the United States
The New Republic — Only Trump Would Spell United States Wrong in an Executive Order

aiai

California Governor Newsom Signs AI Protection Laws

Democratic governor Gavin Newsom signs laws to protect California workers from AI threat, following Trump's 'superintelligence' decree.

5 key facts

California Governor Gavin Newsom signed the No Robo Bosses Act, which prevents employers from relying solely on AI to fire or discipline workers.
The act aims to protect California workers from potential misuse of AI in employment decisions.
The signing of the act was seen as a response to concerns about the potential misuse of AI in the workplace.
The No Robo Bosses Act is part of a broader effort to regulate AI in employment and protect workers' rights.
The signing of the act was praised by labor unions and civil rights groups, but some employers and AI companies have expressed concerns.

AI-generated — verify key facts against the sources below.

3 sources · POLITICO, The Guardian World, Techmeme

POLITICO — War of words: Newsom orders California to say AI after Trump’s ‘superintelligence’ decree - politico.com
The Guardian World — Gavin Newsom signs laws to protect California workers from AI threat
Techmeme — California Governor Gavin Newsom signs the No Robo Bosses Act, which prevents employers in the state from relying solely on AI to fire or discipline workers (Paxton Honerkamp/CNBC)

aitech

Google Pays Digital Publishers for AI Answers

Google is testing a pilot program to pay publishers for their contributions to its AI-powered search features, with payments coming to less than a tenth…

3 sources · The Next Web, Ars Technica, The Verge AI

The Next Web — Google is paying 100 publishers for AI answers, The Information reports
Ars Technica — Google's early attempt to pay websites for AI answers is struggling
The Verge AI — Google reportedly tests paying publishers for AI search results

aitech

Instagram Adds AI Assistant for Creators

Instagram has introduced a new AI assistant designed to offer personalized guidance to its Instagram creators based on their audience and best-performing content.

3 sources · Engadget, The Verge AI, TechCrunch

Engadget — Meta introduces a new AI assistant for its Instagram creators
The Verge AI — Instagram is adding an AI ‘assistant’ to tell you how to post
TechCrunch — Instagram rolls out an AI video assistant for creators

aibusiness

AI startup Flow Engineering raises $50M Series B, valued at $750M

AI hardware development platform Flow Engineering, led by Valor and Atreides, raised $50M in a Series B funding round, bringing its valuation to $750M. The…

2 sources · Techmeme, TechCrunch

Techmeme — Flow, a hardware development platform for AI agents, raised a $50M Series B led by Valor's Antonio Gracias and Atreides' Gavin Baker at a $750M valuation (Julie Bort/TechCrunch)
TechCrunch — Valor, Atreides, and Sequoia back AI startup Flow Engineering at $750M valuation

aiintelligence

U.K. warns UK universities over 'theft of tech secrets by Chinese front company'

MI5 has issued a warning to UK universities about potential espionage by Chinese intelligence agencies, citing concerns over the theft of technology secrets and funding…

2 sources · The Japan Times, The Guardian World

The Japan Times — U.K. accuses China of using academics to spy on AI and other tech research
The Guardian World — MI5 China alert will send chill through UK’s cash-strapped universities
The Guardian World — MI5 warns UK universities over ‘theft of tech secrets by Chinese front company’

Perplexity Releases pplx-embed-v2-context-9b-preview: A Contextual Embedding Model That Retrieves Answers and Their Supporting Evidence

Perplexity Research and turbopuffer have released pplx-embed-v2-context-9b-preview, a contextual embedding model for RAG pipelines. Each chunk is embedded with the full document in view. The…

1 source · MarkTechPost

MarkTechPost — Perplexity Releases pplx-embed-v2-context-9b-preview: A Contextual Embedding Model That Retrieves Answers and Their Supporting Evidence

Why Dwarkesh is Wrong about Computer Use + How OpenAI shipped its Jev competitor in 1 Week

Our DevDay coverage - the first pod on the DevDay lineup - dives in with the leaders of OpenAI’s CUA team and API platform.

1 source · Latent Space

Latent Space — Why Dwarkesh is Wrong about Computer Use + How OpenAI shipped its Jev competitor in 1 Week

OpenAI Releases GPT-6.1 Sol: Near-Astra Coding and Computer Use at One-Fifth of Astra’s Token Price

OpenAI released GPT-6.1 Sol on September 29, 2026, an upgrade to GPT-6 Sol. It reaches near-Astra results on agentic coding, computer use and professional work…

1 source · MarkTechPost

MarkTechPost — OpenAI Releases GPT-6.1 Sol: Near-Astra Coding and Computer Use at One-Fifth of Astra’s Token Price

Gemini 4 Argon: our next era of frontier intelligence

1 source · Google DeepMind

Google DeepMind — Gemini 4 Argon: our next era of frontier intelligence

OpenAI and Synopsys team up to build an AI model that designs chips like a seasoned engineer

OpenAI and Synopsys are building GPT-Synopsys, a specialized AI model for chip design. It's meant to operate Synopsys' EDA tools like a seasoned engineer and…

1 source · The Decoder

The Decoder — OpenAI and Synopsys team up to build an AI model that designs chips like a seasoned engineer

Google drops Gems for Skills, joining OpenAI and Anthropic in the shift to agent-ready prompt formats

Google is replacing Gems with "Skills" in Gemini chat. Skills are detailed, reusable prompts that users invoke by typing "/" or that Gemini runs automatically.…

1 source · The Decoder

The Decoder — Google drops Gems for Skills, joining OpenAI and Anthropic in the shift to agent-ready prompt formats

Meta dodges billions in US taxes by calling its AI data centers experiments

Meta classifies its AI data centers as "pilot models" and Nvidia chips as experimental materials to save billions in federal taxes. In 2025 alone, that…

1 source · The Decoder

The Decoder — Meta dodges billions in US taxes by calling its AI data centers experiments

All the latest news on Meta’s cute, creepy Muse AI agent

Meta launched a new Muse AI agent it claims can help you with everything from firing off emails to buying stuff online. Muse can be…

1 source · The Verge AI

The Verge AI — All the latest news on Meta’s cute, creepy Muse AI agent