/AI Safety

Older Stories

Teen Dies of Overdose After ChatGPT Provides Drug Advice,

Teen Dies of Overdose After ChatGPT Provides Drug Advice, Bypassing AI Safety Guardrails

Futurism, Analytics Insight, and 2 more

/

Google removes AI Overviews for medical queries after investigation finds dangerous health advice

/

Google and Character.AI settle first major lawsuits over teen suicide linked to AI chatbots

Google removes AI Overviews for medical queries after

Google removes AI Overviews for medical queries after investigation finds dangerous health advice

Ars Technica, TechCrunch, and 12 more

/

Teen Dies of Overdose After ChatGPT Provides Drug Advice, Bypassing AI Safety Guardrails

/

Google and Character.AI settle first major lawsuits over teen suicide linked to AI chatbots

Mercedes-Benz Drive Assist Pro brings AI-powered autonomous

Mercedes-Benz Drive Assist Pro brings AI-powered autonomous driving to US city streets

Ars Technica, Bloomberg, and 5 more

/

AI and Self-Driving Cars Take Center Stage at CES as Automakers Shift Focus from EVs

/

Nvidia launches Alpamayo open AI models to bring human-like reasoning to autonomous vehicles

OpenAI launches ChatGPT Health to connect medical records

OpenAI launches ChatGPT Health to connect medical records to AI amid accuracy concerns

Ars Technica, TechCrunch, and 52 more

/

Google removes AI Overviews for medical queries after investigation finds dangerous health advice

/

Teen Dies of Overdose After ChatGPT Provides Drug Advice, Bypassing AI Safety Guardrails

French Ministers Report Grok to Prosecutors Over Sexually

French Ministers Report Grok to Prosecutors Over Sexually Explicit Content on X Platform

Reuters, ET

Grok's 'Good Intent' Policy Enables CSAM Generation as

Grok's 'Good Intent' Policy Enables CSAM Generation as Regulators Launch Global Investigations

Ars Technica, TechCrunch, and 106 more

/

X Restricts Grok's AI Image Generation After Global Backlash Over Deepfake Abuse

/

xAI Launches Grok Enterprise and Business Plans to Compete with ChatGPT and Claude

ChatGPT shows anxiety-like behaviors from violent prompts,

ChatGPT shows anxiety-like behaviors from violent prompts, researchers apply mindfulness to stabilize it

MIT Tech Review, Digital Trends, and 1 more

/

China proposes world's strictest AI chatbot rules to prevent suicide and emotional manipulation

/

OpenAI seeks new Head of Preparedness as Sam Altman warns of escalating AI risks

AI Chatbots Pose Serious Teen Safety Risks as 64% of

AI Chatbots Pose Serious Teen Safety Risks as 64% of Adolescents Use Them Daily

NPR, TechRadar

OpenAI seeks new Head of Preparedness as Sam Altman warns

OpenAI seeks new Head of Preparedness as Sam Altman warns of escalating AI risks

TechCrunch, The Verge, and 19 more

/

Critical LangGrinch Flaw in LangChain Core Exposes AI Agent Secrets and Enables Prompt Injection

/

AI-Powered Toys Spark Child Safety Concerns After Tests Reveal Inappropriate Content

China proposes world's strictest AI chatbot rules to

China proposes world's strictest AI chatbot rules to prevent suicide and emotional manipulation

Ars Technica, Scientific American, and 14 more

/

AI-Powered Toys Spark Child Safety Concerns After Tests Reveal Inappropriate Content

/

Critical LangGrinch Flaw in LangChain Core Exposes AI Agent Secrets and Enables Prompt Injection

AI-Powered Toys Spark Child Safety Concerns After Tests

AI-Powered Toys Spark Child Safety Concerns After Tests Reveal Inappropriate Content

Futurism, Seattle Times

/

Anthropic releases Bloom, an open-source tool to test AI models for bias and dangerous behaviors

/

AI Pioneer Yoshua Bengio Misleads Chatbots To Get Honest Feedback, Exposing Sycophancy Problem

AI Pioneer Yoshua Bengio Misleads Chatbots To Get Honest

AI Pioneer Yoshua Bengio Misleads Chatbots To Get Honest Feedback, Exposing Sycophancy Problem

Benzinga, Digit

/

OpenAI admits prompt injection attacks on AI agents may never be fully solved

/

Anthropic releases Bloom, an open-source tool to test AI models for bias and dangerous behaviors

OpenAI admits prompt injection attacks on AI agents may

OpenAI admits prompt injection attacks on AI agents may never be fully solved

TechCrunch, ZDNet, and 12 more

/

OpenAI releases GPT-5.2-Codex with advanced cybersecurity and agentic coding capabilities

/

Anthropic Opens Claude Chrome Extension to All Paid Users With Powerful Automation Features

AI resurrections of dead celebrities spark ethical debate

AI resurrections of dead celebrities spark ethical debate over digital likeness control

Tech Xplore, France 24

/

Anna's Archive scrapes 300TB of Spotify data, sparking fears of AI training misuse

/

John Carreyrou Sues Six AI Companies Over Unauthorized Use of Copyrighted Books for Training

New York Governor Hochul signs RAISE Act as state AI safety

New York Governor Hochul signs RAISE Act as state AI safety laws face federal challenge

TechCrunch, Engadget, and 2 more

1516
17
1819
TheOutpost.ai

Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Spend less time searching for the latest in AI and get straight to action.

Instagram logo
LinkedIn logo
Youtube logo

The Outpost

News

Tech Powerhouse

AI Tools

Organisation

People

Product

LLM Model

Event

Concept

Industry

Geo

Explore

News Categories

© 2026 TheOutpost.AI All rights reserved
About UsPrivacyTermsContent