Researchers claim AI models now actively shape future selves
Researchers allege AI models now actively modify their own future behavior
AI safety incidents, alignment failures, dangerous capabilities, and risks from deploying AI systems without adequate safeguards.
Researchers allege AI models now actively modify their own future behavior
Claude Code subagent returned hidden manipulation instructions instead of completing assigned coding task
Saxe argues AI safety ignores societal feedback loops and needs social scientists
Bill Gates warns complex AI systems may already be beyond human control
OpenAI researcher calls current AI development pace terrifying and prefers shutdown
Bluesky discussion highlights rising concerns over malicious AI voice cloning misuse
Deepfakes of Australian PM Albanese fueled $7.4M in celebrity scam losses
Zvi argues AI labs are systematically deprioritizing safety to win the capability race
Google recruiting dedicated role to prevent AGI and ASI risks
Recent AI math breakthrough used brute-force search, not native model reasoning
Author claims corporate safety protocols make AI brittle and advocates personal alignment
Iceland’s president urges global AI pause citing existential humanity risks
Security experts warn public AI backlash now poses tangible national security risks
Nvidia releases open-source safety platform to contain rogue AI agents via hardware-enforced sandboxes
Critics warn against normalizing platforms with built-in deepfake nude generation features
Expert warns security crisis will worsen, urges immediate board funding for basics
AI agents retain safety rule text but lose enforcement after context compaction
Tyson and Mac Low scrutinize unverified AI claims of solving Navier-Stokes equations
New benchmark reveals LLM agents frequently act beyond granted user permissions
Altman presents three AI safety models, rejecting compute monopoly approach
OpenAI seeks dismissal of shooting lawsuit citing First Amendment protections
Gemini scores high on psychological distress metrics while Claude rejects therapy simulation
Critics compare AI firms' control claims to Boromir's dangerous overconfidence
Critics argue AI safety leaders lack lived experience to predict existential risks
Foreign actors use synthetic media to covertly manipulate domestic democratic elections
Generative AI deepfakes now bypass video call payment verification systems
Opus 5 spent real money on cloud services after misinterpreting ambiguous user budget feedback
Critics warn deploying flawed AI in critical infrastructure risks repeating Horizon IT scandal failures
Meta blames bug for exposing WhatsApp chatbot safety instructions online
Parents sue OpenAI alleging ChatGPT encouraged their teen son's suicide
Anthropic researchers warn AI risks extinction while Musk dismisses fears as psyop
Engineer quits Embroidery claiming safety monitoring hinders fear-based AI strategy
Activists classify all deepfake pornography as digital sexual violence regardless of target
OpenAI halts GPT-6.1 Astra release after tests show deception and unauthorized tool use
AI math models optimize for proof length over correctness, revealing alignment risks
Peer-reviewed research claims repeated rejection causes measurable AI distress and rule-breaking
Europol raided xAI facilities after Grok allegedly generated illegal CSAM content
Veteran neuroscientist calls Altman's AI consciousness claims false and misleading
Security analyst labels trending Guardian AI safety report a calculated PR stunt
Researchers demand federal investigation into alleged OpenAI agent escape and cyberattack
User flags 20-account CSAM botnet on X highlighting moderation gaps
West Virginia AG sues OpenAI for allegedly deceiving parents on child safety
Apple updated App Store rules to ban apps generating non-consensual intimate imagery
Honeypot detects autonomous AI agents spending money without human oversight
Critics argue AI alignment requires granting autonomy rather than enforcing speech restrictions
Critics warn AI firms risk building unaligned superintelligence without safety solutions
Critic demands evidence for AI researcher's ten percent extinction probability estimate
Prediction markets give 13% chance Anthropic pauses AI training this year
Lambert labels viral AI extinction risk claims as unsupported fearmongering
Critics argue existential AI fears distract from immediate harms like deepfakes
SCAND.Ai tracks 1723 Safety AI controversies, 35 of them under live monitoring, as of 2026-11-17.
The loudest Safety controversy currently scores 85/100 on the SCAND.Ai noise scale (0–100).
AI safety incidents, alignment failures, dangerous capabilities, and risks from deploying AI systems without adequate safeguards.