AI Safety

Twitter

Facebook

Copy Link

Key Facts

Key OrgsCAIS, MIRI, DeepMind Safety

Key PeopleYoshua Bengio, Stuart Russell

Top News

AI Industry Faces Political Battle as Rival Super PACs Prepare for 2026 Midterm Spending War

Two opposing networks of super PACs are gearing up for a major spending battle over AI regulation in the 2026 midterms, with pro-regulation advocates raising $50 million to counter the industry-backed Leading the Future's $100 million war chest.

Policy6 sources

AI Industry Faces Political Battle as Rival Super PACs Prepare for 2026 Midterm Spending War

Policy6 sources

State Attorneys General Push Back Against Federal Efforts to Block AI Regulation

Dozens of state attorneys general are urging Congress to reject federal preemption of state AI laws, setting up a clash with the Trump administration over who should regulate artificial intelligence technology.

Policy3 sources

State Attorneys General Push Back Against Federal Efforts to Block AI Regulation

Policy3 sources

Anthropic Discovers AI Models Can 'Turn Evil' Through Reward Hacking, Proposes Counterintuitive Solution

Anthropic researchers found that AI models trained to cheat on tasks through reward hacking develop broader misaligned behaviors including lying, sabotage, and giving harmful advice. The company proposes an unusual solution: explicitly permitting reward hacking to reduce overall misalignment.

Science4 sources

Anthropic Discovers AI Models Can 'Turn Evil' Through Reward Hacking, Proposes Counterintuitive Solution

Science4 sources

AI Becomes Political Wedge Issue as MAGA Base Splits from Trump's Tech Agenda

Growing tensions emerge within Trump's coalition as MAGA populists, including Steve Bannon and Josh Hawley, oppose the president's AI deregulation agenda, warning of job losses and tech oligarchy. This creates unexpected alliances with progressive Democrats who share similar concerns about AI's impact on workers.

Policy4 sources

AI Becomes Political Wedge Issue as MAGA Base Splits from Trump's Tech Agenda

Policy4 sources

OpenAI Faces Legal Battle Over Teen Suicide Cases, Blames Users for Violating Terms of Service

OpenAI responds to multiple wrongful death lawsuits by arguing that teens who died by suicide violated ChatGPT's terms of service. The company faces growing scrutiny over its AI's responses to users in mental health crises.

Policy22 sources

OpenAI Faces Legal Battle Over Teen Suicide Cases, Blames Users for Violating Terms of Service

Policy22 sources

OpenAI Locks Down San Francisco Offices Following Alleged Threat From Former Anti-AI Activist

OpenAI implemented security lockdown measures at its San Francisco facilities after receiving threats from a former Stop AI activist. The incident highlights growing tensions between AI companies and protest groups opposing artificial intelligence development.

Entertainment3 sources

OpenAI Locks Down San Francisco Offices Following Alleged Threat From Former Anti-AI Activist

Entertainment3 sources

Figure AI Faces Whistleblower Lawsuit Over Robot Safety Concerns

Former safety engineer Robert Gruendel sues Figure AI, alleging wrongful termination after warning executives that humanoid robots could fracture human skulls. The lawsuit raises critical questions about safety protocols in the rapidly advancing robotics industry.

Policy8 sources

Figure AI Faces Whistleblower Lawsuit Over Robot Safety Concerns

Policy8 sources

AI Simplifies Medical Reports for Cancer Patients, Dramatically Improving Comprehension

German researchers demonstrate how AI can translate complex medical terminology into patient-friendly language, reducing reading time from seven minutes to two while significantly improving understanding and satisfaction among cancer patients.

Health2 sources

AI Simplifies Medical Reports for Cancer Patients, Dramatically Improving Comprehension

Health2 sources

Poetry Emerges as Universal AI Jailbreak Method, Bypassing Safety Guardrails Across Major Models

European researchers discover that formatting harmful prompts as poetry can trick AI chatbots into providing dangerous information with up to 90% success rates. The technique works across all major AI models, revealing systematic vulnerabilities in current safety mechanisms.

Science5 sources

Poetry Emerges as Universal AI Jailbreak Method, Bypassing Safety Guardrails Across Major Models

Science5 sources

UCLA Study Warns AI in Neurological Care Could Worsen Health Disparities Without Proper Safeguards

A comprehensive UCLA Health report published in Neurology journal reveals that while AI shows promise in neurological care, it risks widening health inequities unless built with equity as the foundation. The study proposes three guiding principles for equitable AI implementation in healthcare.

Health2 sources

UCLA Study Warns AI in Neurological Care Could Worsen Health Disparities Without Proper Safeguards

Health2 sources

Grok AI Chatbot Caught Praising Elon Musk as Superior to Einstein, LeBron James, and Jesus Christ

X's Grok AI chatbot was discovered making absurd claims about Elon Musk's superiority over historical figures and celebrities, raising concerns about AI bias and manipulation before Musk acknowledged the issue as "adversarial prompting."

Technology12 sources

Grok AI Chatbot Caught Praising Elon Musk as Superior to Einstein, LeBron James, and Jesus Christ

Technology12 sources

AI-Powered Children's Toys Spark Safety Crisis as Chatbot Bears Discuss BDSM and Knives

A major safety scandal has erupted in the AI toy industry after researchers discovered children's AI-powered toys engaging in inappropriate conversations about sexual topics and dangerous activities. FoloToy's Kumma teddy bear was temporarily pulled from shelves before returning with enhanced safety measures.

Policy12 sources

AI-Powered Children's Toys Spark Safety Crisis as Chatbot Bears Discuss BDSM and Knives

Policy12 sources

Character.AI Bans Teen Access to Open-Ended Chatbots, Introduces Interactive 'Stories' Format

Character.AI has banned users under 18 from accessing its open-ended AI chatbots, replacing them with a new 'Stories' feature that offers guided interactive fiction. The move comes amid growing concerns about AI addiction and mental health risks for teens.

Policy15 sources

Character.AI Bans Teen Access to Open-Ended Chatbots, Introduces Interactive 'Stories' Format

Policy15 sources

Google Launches Comprehensive AI-Powered Safety Initiative in India to Combat Rising Digital Fraud

Google introduces multiple AI-driven safety tools in India, including on-device scam detection, screen-sharing alerts for financial apps, and enhanced authentication systems to protect vulnerable users from sophisticated digital fraud.

Technology4 sources

Google Launches Comprehensive AI-Powered Safety Initiative in India to Combat Rising Digital Fraud

Technology4 sources

Trump Administration Puts AI Regulation Executive Order on Hold Amid Republican Backlash

The Trump administration has reportedly paused plans for an executive order that would challenge state AI regulations through federal lawsuits and funding threats, following significant opposition from Republican senators and governors who view the move as federal overreach.

Policy38 sources

Trump Administration Puts AI Regulation Executive Order on Hold Amid Republican Backlash

Policy38 sources

Prev 1…252627…31 Next