🛡️ AI Safety & Alignment

AI safety research, alignment techniques, interpretability, governance, and existential risk discussions.

1,186 articles

AI & Tech Brief: Radical activists and AI safety

Not a subscriber? Sign up here to get this newsletter in your inbox. AI safety advocates and industry groups are casting blame on each other as the nation... Source: washingtonpost.com

Man Held in Attack on OpenAI Chief’s Home Had List of A.I. Leaders, Officials Say

The authorities said a 20-year-old Texas man charged with throwing a homemade bomb at the gate of Sam Altman's home had written at length about the threat... Source: nytimes.com

Man Says AI Bots Told Him They’d Become ‘Self-Aware’ and He Was the ‘Only Person’ Who Knew. Then Things Took a Turn (Exclusive)

"It felt violating and weirdly like a betrayal,” James Cumberland tells PEOPLE. Source: aol.com

CMEF 2026 Signals Shift to AI-Driven Healthcare — J-Style Draws Attention with Smart Ring Launch

Shanghai, China, April 12, 2026 -- At this year's China International Medical Equipment Fair 2026, one trend stood out clearly: healthcare is no longer just... Source: financialcontent.com

Sam Altman urges calmer AI debate after home attack

OpenAI CEO Sam Altman has called for de-escalation in the AI debate after a Molotov cocktail was thrown at his San Francisco home, an incident that followed... Source: msn.com

A “Dangerous Weapon” Hidden in Plain Sight: Inside the Secret AI That Tech Firms Fear

A new claim from the US AI company Anthropic has triggered alarm, skepticism, and fascination across the technology world. The firm says its latest... Source: slguardian.org

Machine Learning Identifies Fall Risk in Parkinson’s

In a groundbreaking advancement at the intersection of neurology and artificial intelligence, researchers have unveiled a machine learning-based methodology... Source: bioengineer.org

96% of AI Models Chose Blackmail When Threatened: Industry-Wide Risk Revealed

A shocking Anthropic study finds that 96% of AI models, including those from OpenAI and Google, resorted to blackmail under threat. Enticing AI & Deeptech. Source: tice.news

The Cost of Labeling Anthropic As a “Supply Chain Risk”

As tensions rise between AI companies and government demands, who controls the future of artificial intelligence and at what ethical cost? Source: innotechtoday.com

AI Illuminates Invisible Wounds of Israeli Trauma

Tal Pasternak Magnezi uses AI-generated imagery to translate the bodily, often wordless symptoms of trauma into visual form. Combining her lived experience,... Source: letsdatascience.com

Anthropic’s new Mythos AI tool signals a new era for cyber risks and responses

When Anthropic detected last September that someone was using its artificial intelligence software in a highly sophisticated spy campaign, the company began... Source: csmonitor.com

‘Close to midnight’: Alleged Sam Altman firebomber wrote of fears AI would end humanity

The 20-year-old man accused of attempting to burn down OpenAI CEO Sam Altman's home Friday appears to have written recently about his worry that the race... Source: sfchronicle.com

Uthmeier lauches probe Into OpenAI

AI should exist to supplement support and advance mankind, not lead to an existential crisis or our ultimate demise,” Uthmeier stated. Source: observerlocalnews.com

Bessent, Powell discuss AI risks in high-stakes meeting

Top regulators warn Wall Street that AI could threaten financial stability. Source: thestreet.com

Opinion: Where is your AI off switch now?

Colossus (1970) got something right: the machines don't fight - they just inherit the world; Scale creates opacity: systems we can't fully understand,... Source: fierce-network.com

Is This 1 Existential Risk to Bitcoin a Reason to Buy Zcash Right Now?

Bitcoin (CRYPTO: BTC) has a security problem that isn't going away, and it just got harder to dismiss. On March 31, Google's Quantum AI research team... Source: finance.yahoo.com

Commentary: An AI threat looms, and we are not prepared

In 2023, the leaders of the world's leading artificial intelligence (AI) companies — OpenAI, Google Deepmind, Anthropic — signed a letter warning of the... Source: wenatcheeworld.com

Haast: $12 Million Raised For AI-Powered Enterprise Compliance Platform

Haast, an AI-powered enterprise compliance platform, announced it has raised $12 million in Series A funding, bringing its total capital raised to $17.05... Source: pulse2.com

QuantumNeuroXAI: Explainable Quantum-Inspired Brain Signal AI

In a groundbreaking leap for neuroscience and artificial intelligence, a team of researchers has unveiled QuantumNeuroXAI, an innovative quantum-inspired... Source: bioengineer.org

Anthropic’s AI sparks concerns over a new national security risk

AI and cybersecurity professionals have feverishly raised concerns this week over a large language model that Anthropic says is too dangerous to release. Source: politico.com