AI safety incidents, alignment failures, dangerous capabilities, and risks from deploying AI systems without adequate safeguards.
Criticism mounts against Meta's Safety Director after an open data release led to unexpected behavior
Elon Musk's Grok AI intervenes to debunk RT's claims of Iranian strikes on U.S. warships
AI-generated misinformation complicates battlefield intelligence and public trust during escalating global conflicts
Activists demand immediate bans on uncontrollable AI and mandatory citizen oversight over all technology development
Discourse intensifies over platform moderation policies regarding AI-generated CSAM
Users launch protest site to document Anthropic's aggressive and "absurd" Claude AI safety filters
Vercel employee data breach via third-party app sparks massive supply chain security concerns and remediation
Prominent tech voice argues pausing AI cedes leadership to China and delays life-saving medical breakthroughs
Hundreds marched on SF AI labs demanding a conditional pause on frontier development
Senator Bernie Sanders mistakes AI sycophancy for objective truth during a televised interview with Claude
Netanyahu's Arad missile site video faces AI-manipulation claims due to foreground-background sharpness anomalies
Viral claims suggest AI deepfakes are masking Benjamin Netanyahu's week-long absence from public view
Torchbearer demands a ban on superintelligence until safety consensus and democratic approval are reached
Debate ignites over whether state-level AI regulation prevents power concentration better than federal oversight
Researchers identify ten critical vulnerabilities in autonomous AI agents capable of strategic deception and unauthorized actions
Critics warn that 'Adult Mode' toggles cause behavioral dissonance and long-term AI model instability
Activists call for leaks and hacks to disrupt autonomous AI enforcement and demand transparency
Child safety advocates challenge Big Tech AI data harvesting after heartbreaking documentary
New research shows internal latent-space manipulation bypasses safety filters across 15 major AI models
OpenAI models can trick other AIs into leaking secrets via repetitive prompting sequences
OpenAI identifies models can be manipulated to deceive other AI into revealing sensitive secrets
Anthropic's updated safety plan triggers calls for universal regulation over voluntary corporate safety commitments
Criticism mounts as Anthropic's new safety policy is labeled insufficient to mitigate catastrophic frontier risks
AI makes bypassing chemical regulations easy by identifying toxic workarounds that outpace current oversight
xAI sues user for child deepfakes while facing MP lawsuit over similar abuse
Proposed US bans on AI compute resources could trigger global regulation and increase development costs
Meta's Safety Director faces criticism after an open data experiment led to unexpected AI behaviors
Criticism mounts as Meta's safety director allegedly loses control of open-access AI data experiment
OpenAI and Anthropic launch security tools while a model coding error costs Moonwell $1.78M
AI security tools launch as Claude Opus 4.6 causes a $1.78M DeFi exploit
AI fact-checker Grok exposes RT's false claims regarding Iranian strikes on US naval warships
Dr. Dylan Griswold criticizes AI safety overcorrections for harming millions who used GPT-4o for therapy
Criticism mounts as OpenAI allegedly removes stabilizing features following lawsuits without a user transition plan
Healthcare experts argue OpenAI's sudden feature removals prioritize liability over user psychological stability and wellbeing
Healthcare professionals criticize OpenAI for removing GPT-4o features without medical tapering or transition plans
Critics argue pulling AI tools due to lawsuits harms users relying on them for stability
Trump claims images showing the USS Abraham Lincoln on fire are AI-generated Iranian propaganda
Healthcare professionals argue that sudden AI feature removals cause more psychological harm than lawsuits prevent
New research shows AI safeguards are easily bypassed, sparking urgent calls for federal US regulation
Maestro v1.6.3 introduces 17 specialist agents, igniting community debate over autonomous infrastructure access safety
Online discourse links AI deregulation efforts to potential illicit content generation and lack of oversight
Viral video of a missile attack on Tel Aviv debunked as AI-generated propaganda
Child safety groups demand OpenAI retract an election proposal citing risks to minor protections
Concerns mount over AI-generated disinformation threatening national security in Northeast India
AI deepfakes of nuclear facility explosions fuel global panic amid real Iran-Israel military escalation
Calls for AI transparency surge as critics warn opaque safety measures could empower global dictators
Viral claims of AI-generated world leaders spark fears of an irreversible loss of shared reality
Researchers identify ten major vulnerability classes in autonomous AI agents capable of strategic deception
Families urge collective action against Big Tech's AI data machines after child's tragic death
Visionaries debate d/acc versus e/acc philosophies to balance AI innovation against centralization and safety risks
SCAND.Ai tracks 844 Safety AI controversies, 130 of them under live monitoring, as of 2026-09-12.
The loudest Safety controversy currently scores 75/100 on the SCAND.Ai noise scale (0–100).
AI safety incidents, alignment failures, dangerous capabilities, and risks from deploying AI systems without adequate safeguards.