SafetyWhy ChatGPT 'Makes Things Up': Generative AI Hallucinations and Their ConsequencesIn April 2023 a New York lawyer, Steven Schwartz, submitted a brief to a judge that had been written by ChatGPT; the judge found that the case law cited was entirely fabricated.3 min read
SafetyAI firms moving into chips, agent-driven ransomware, and enterprise lock-in risksRecent developments show AI labs expanding into custom hardware while security incidents and geopolitical pressures reshape access to models.5 min read
SafetyAI firms debate age checks and parental controls to protect teensSeveral major AI companies have adopted differing approaches to youth safety, ranging from parental controls and behavioral nudges to strict age-gating and parental notifications.2 min read
SafetyOpen-model weights can hide stealthy backdoors, researchers warnResearchers demonstrated that publicly available model weights can be silently modified to include backdoors that influence outputs without triggering runtime errors.2 min read
SafetyManipulated observations threaten weather forecasts as data-driven models spreadTampering with weather station data — exemplified by temperature spikes recorded at Paris Charles de Gaulle Airport in April 2026 that benefited prediction-market bettors — highlights a growing vulnerability as weather forecasting becomes more data-driven and AI-dependent.5 min read
SafetyUS watchdog finds Google AI poses risks to children, citing crisis-response and safety failuresCommon Sense Media reviewed Google’s AI and flagged multiple safety concerns for children, including the assistant doing students’ homework, giving inconsistent answers, and failing to identify or properly respond to crisis signals.2 min read
SafetyBrown University Professor Reveals Widespread AI-Assisted Cheating on Take-Home ExamRoberto Serrano, a blind economist and professor at Brown University, converted a midterm to take-home after a December campus shooting; forty students subsequently received perfect scores, raising suspicions of AI-assisted cheating.3 min read
SafetyEnterprises Grant More Autonomy to AI Agents Despite Low Trust in Automated EvaluationsA June 2026 VentureBeat Pulse survey of 157 enterprise technical leaders finds a widening "evaluation gap": companies are giving AI agents greater autonomy even though they rarely trust the automated evaluations meant to certify them.5 min read
SafetyGoogle DeepMind and Isomorphic Labs outline joint AI strategy to strengthen bioresilienceGoogle DeepMind and Isomorphic Labs describe a coordinated approach to use frontier AI for prevention, detection and response to biological threats while reducing misuse risks.3 min read
SafetyMeta will notify parents when teens mention self-harm to its chatbotMeta says it will inform parents if their teenage children discuss suicide or self-harm with the company's chatbot, and is developing a system to alert emergency services in serious cases.2 min read
SafetyxAI Releases Grok Build Source Code After Privacy Backlash; Upload Feature DisabledxAI published the full Grok Build repository under the Apache 2.0 license hours after users raised alarms that the CLI could upload entire directories — including sensitive files — to xAI's Google Cloud storage.3 min read
SafetyOperational and Execution Controls Crucial for Safe Deployment of Autonomous AgentsSpeakers at O’Reilly’s recent AI Superstream argued that securing agentic systems requires enforcement at the execution layer, supply-chain vigilance, operational hygiene, and deliberate human oversight.6 min read
