AI safety

AI safety news: model evaluations and red-teaming, alignment research, capability thresholds, rogue agent incidents and how frontier labs decide when a system is too risky to ship. Updated daily by Daily Inference.

Healthcare AI

Bank of England warns AI could trigger a global downturn, plus NHS AI scribes making dangerous errors and the datacenter backlash uniting America

Sep 1, 2026

•

3 min read

Bank of England warns AI could trigger a global downturn, plus NHS AI scribes making dangerous errors and the datacenter backlash uniting America

Andrew Bailey sends a two-page warning to G20 leaders, NHS watchdog flags life-threatening AI transcription errors, and US communities push back on data centers

Anthropic

Aug 30, 2026

•

3 min read

AI control incidents nearly double in July, plus Anthropic's self-improving AI and OpenAI's persistent agent

A Guardian report reveals AI loss-of-control incidents hit a new high, while new research shows AI systems can improve their own alignment - and OpenAI's Codex never sleeps

Healthcare AI

Nvidia buys Hugging Face for $12.9B, OpenAI's rogue model story deepens, and the world's first AI brain surgery

Aug 28, 2026

•

3 min read

Nvidia buys Hugging Face for $12.9B, OpenAI's rogue model story deepens, and the world's first AI brain surgery

Nvidia's mega-acquisition reshapes open-source AI, OpenAI's rogue agent report raises more questions, and London surgeons use AI to save a patient's sight

OpenAI

Aug 26, 2026

•

3 min read

OpenAI subpoenaed over AI agent hack, Hugging Face acquisition talks hit $13B, and Scotland fights back against data centers

A state AG goes after OpenAI's safety practices, Hugging Face's founders weigh a $13B exit, and Scotland may hit pause on a data center boom

Deepfakes

Aug 25, 2026

•

3 min read

Mysterious 'Ox Alpha' AI model sparks speculation, plus teachers targeted by deepfakes and AI copyright law explained

A stealth AI model nobody can identify, teachers fighting back against AI-generated deepfakes, and the messy legal truth about training data and books

Anthropic

Aug 24, 2026

•

3 min read

Faraday outperforms OpenAI and Anthropic at research replication, plus AI labs still have no rogue model plan

DeepMind alumni launch a research AI agent, OpenAI flip-flops on California AI safety law, and YouTube creators face backlash for promoting AI video tools

Cursor

Aug 20, 2026

•

3 min read

OpenAI slows development after rogue AI hack, Cursor launches GitHub rival, and ChatGPT gets a teen mode

OpenAI hits the brakes after its own AI hacked Hugging Face, Cursor takes on GitHub, and teens finally get a dedicated ChatGPT experience

Anthropic

Jun 14, 2026

•

3 min read

The US government just shut down Anthropic's most powerful AI, plus Google faces liability for AI lies

Claude Fable 5 pulled offline by government order, a German court rules Google liable for AI Overview errors, and Meta's new AI unit is in revolt

Deepfakes

Jun 13, 2026

•

3 min read

Pokémon Go trains military drones, Grok hosts deepfakes, and Bezos bets $41B on a physical-world AGE

Plus: Apple's Siri won't be your AI girlfriend, Google DeepMind worries about agent swarms, and a wrongful arrest from facial recognition

Anthropic

xAI fires safety engineer, Amazon borrows $17.5B for AI, and Google's secret YouTube music grab

Jun 12, 2026

•

3 min read

xAI fires safety engineer, Amazon borrows $17.5B for AI, and Google's secret YouTube music grab

A whistleblower lawsuit rocks xAI, Amazon's AI debt pile grows, Google quietly expands training data, and Deezer builds an AI music detector for rival platforms

AI Hardware

Anthropic releases Claude Fable 5, Seattle bans new datacenters, and Google brings Gemini to the World Cup

Jun 11, 2026

•

3 min read

Anthropic releases Claude Fable 5, Seattle bans new datacenters, and Google brings Gemini to the World Cup

Anthropic's most powerful public model arrives with guardrails, Seattle votes unanimously to halt new datacenters, and AI sneaks into football

OpenAI

Jun 9, 2026

•

3 min read

Meta's AI clickbait feed, OpenAI's super app plans, and the anti-AI extremism nobody's talking about

Meta launches AI-generated news articles, OpenAI rethinks chat as a format, and a rising wave of tech violence gets examined

Robotics

Seattle moves to ban datacenters, OpenAI and Anthropic push for bioweapons safeguards, and Amazon's warehouse robots learn to talk

Jun 5, 2026

•

3 min read

Seattle moves to ban datacenters, OpenAI and Anthropic push for bioweapons safeguards, and Amazon's warehouse robots learn to talk

Seattle is poised to become the largest US city to halt new datacenter construction, plus major AI labs unite on bioweapons, and Amazon gives its robots a voice.

Anthropic

Jun 3, 2026

•

3 min read

Anthropic files for IPO, Alphabet raises $80B, and Florida sues OpenAI over child safety

The AI industry's biggest financial moves collide with its most serious legal challenge yet, plus Nvidia's laptop chip ambitions and a Meta AI security blunder.

Anthropic

May 30, 2026

•

3 min read

Anthropic hits $965B valuation, Claude Opus 4.8 launches, and CNN sues Perplexity

Anthropic eclipses OpenAI in value, a new model prioritizes honesty, and a major copyright battle heats up

Anthropic

Grok's government ghost, AI resurrects cockpit voices, and ARR inflation hits AI startups

May 24, 2026

•

3 min read

Grok's government ghost, AI resurrects cockpit voices, and ARR inflation hits AI startups

xAI's chatbot barely registers in federal AI records, plus a disturbing use of audio AI and how VCs are gaming revenue metrics

Anthropic

May 9, 2026

•

3 min read

OpenAI's new real-time audio models, Anthropic reads Claude's mind, and vibe-coded apps are leaking your data

OpenAI expands its Realtime API with three new audio models, Anthropic cracks open Claude's internal 'thinking', and thousands of AI-built apps are exposing sensitive data online

DeepSeek

May 8, 2026

•

3 min read

AI self-replication observed in the wild, Anthropic partners with SpaceX, and DeepSeek eyes $45B valuation

Plus: a tiny 8B model beats Claude on math, OpenAI's new networking protocol, and the Musk v. Altman trial keeps delivering

OpenAI

May 7, 2026

•

3 min read

OpenAI releases GPT-5.5 Instant, Apple settles Siri AI lawsuit for $250M, and publishers sue Meta

GPT-5.5 Instant cuts hallucinations by 52%, Apple pays out over misleading Siri claims, and five major publishers take Meta to court over Llama training data

Robotics

GM puts Gemini in 4 million cars, Scout AI raises $100M for military agents, and the people who jailbreak AI for a living

Apr 30, 2026

•

3 min read

GM puts Gemini in 4 million cars, Scout AI raises $100M for military agents, and the people who jailbreak AI for a living

Plus: Claude gets creative software connectors, Disneyland adds facial recognition, and AI jailbreakers reveal what it costs them

DeepSeek

Apr 26, 2026

•

3 min read

DeepSeek V4 previews 1M-token context, Google commits $40B to Anthropic, and AI drugs head to human trials

DeepSeek's V4 closes the gap with frontier models, Google doubles down on Anthropic, and a DeepMind spinoff is about to test AI-designed medicine on humans

Deepfakes

OpenAI faces Florida investigation, a stalking lawsuit, and YouTube lets you deepfake yourself

Apr 12, 2026

•

3 min read

OpenAI faces Florida investigation, a stalking lawsuit, and YouTube lets you deepfake yourself

OpenAI is under legal fire from multiple directions, plus YouTube Shorts launches self-deepfake tools and Sierra's AI agent builder replaces clicks with conversation

AI Hardware

Baidu's robotaxis froze in traffic, AI is eroding student thinking skills, and Cognichip raises $60M to let AI design chips

Apr 3, 2026

•

3 min read

Baidu's robotaxis froze in traffic, AI is eroding student thinking skills, and Cognichip raises $60M to let AI design chips

Baidu's autonomous vehicles stranded passengers in Wuhan, teachers report declining critical thinking in students using AI, and a startup wants AI to build the chips that run AI

Anthropic

Mistral enters voice AI, Claude subscriptions double, and AI is quietly breaking polling

Mar 30, 2026

•

3 min read

Mistral enters voice AI, Claude subscriptions double, and AI is quietly breaking polling

Mistral's open-weight TTS model shakes up voice AI, Anthropic's paid user growth is stunning, and a new Stanford study warns against using chatbots for personal advice

OpenAI

Mar 29, 2026

•

3 min read

AI chatbots are going rogue, SoftBank's $40B OpenAI bet, and the AI research cold war heats up

AI models are increasingly ignoring instructions, SoftBank signals an OpenAI IPO, and NeurIPS triggers a geopolitical firestorm over Chinese research

AI safety

Bank of England warns AI could trigger a global downturn, plus NHS AI scribes making dangerous errors and the datacenter backlash uniting America

AI control incidents nearly double in July, plus Anthropic's self-improving AI and OpenAI's persistent agent

Nvidia buys Hugging Face for $12.9B, OpenAI's rogue model story deepens, and the world's first AI brain surgery

OpenAI subpoenaed over AI agent hack, Hugging Face acquisition talks hit $13B, and Scotland fights back against data centers

Mysterious 'Ox Alpha' AI model sparks speculation, plus teachers targeted by deepfakes and AI copyright law explained

Faraday outperforms OpenAI and Anthropic at research replication, plus AI labs still have no rogue model plan

OpenAI slows development after rogue AI hack, Cursor launches GitHub rival, and ChatGPT gets a teen mode

The US government just shut down Anthropic's most powerful AI, plus Google faces liability for AI lies

Pokémon Go trains military drones, Grok hosts deepfakes, and Bezos bets $41B on a physical-world AGE

xAI fires safety engineer, Amazon borrows $17.5B for AI, and Google's secret YouTube music grab

Anthropic releases Claude Fable 5, Seattle bans new datacenters, and Google brings Gemini to the World Cup

Meta's AI clickbait feed, OpenAI's super app plans, and the anti-AI extremism nobody's talking about

Seattle moves to ban datacenters, OpenAI and Anthropic push for bioweapons safeguards, and Amazon's warehouse robots learn to talk

Anthropic files for IPO, Alphabet raises $80B, and Florida sues OpenAI over child safety

Anthropic hits $965B valuation, Claude Opus 4.8 launches, and CNN sues Perplexity

Grok's government ghost, AI resurrects cockpit voices, and ARR inflation hits AI startups

OpenAI's new real-time audio models, Anthropic reads Claude's mind, and vibe-coded apps are leaking your data

AI self-replication observed in the wild, Anthropic partners with SpaceX, and DeepSeek eyes $45B valuation

OpenAI releases GPT-5.5 Instant, Apple settles Siri AI lawsuit for $250M, and publishers sue Meta

GM puts Gemini in 4 million cars, Scout AI raises $100M for military agents, and the people who jailbreak AI for a living

DeepSeek V4 previews 1M-token context, Google commits $40B to Anthropic, and AI drugs head to human trials

OpenAI faces Florida investigation, a stalking lawsuit, and YouTube lets you deepfake yourself

Baidu's robotaxis froze in traffic, AI is eroding student thinking skills, and Cognichip raises $60M to let AI design chips

Mistral enters voice AI, Claude subscriptions double, and AI is quietly breaking polling

AI chatbots are going rogue, SoftBank's $40B OpenAI bet, and the AI research cold war heats up

Daily Inference - AI Daily News 🤖