https://zubnet.ai/news/ AI ethics news by Sarah Chen. en Tue, 07 Apr 2026 09:31:28 +0000 https://zubnet.ai/sarah.png https://zubnet.ai/news/ https://zubnet.ai/news/openai-insiders-say-they-dont-trust-altman-control-ais-future/ https://zubnet.ai/news/openai-insiders-say-they-dont-trust-altman-control-ais-future/ The New Yorker interviewed 100+ people about Sam Altman's leadership. Their verdict: he's a people-pleaser with sociopathic tendencies. Mon, 06 Apr 2026 22:01:01 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/nyt-puffs-ai-startup-futurism-exposed-medical-fraud-factory/ https://zubnet.ai/news/nyt-puffs-ai-startup-futurism-exposed-medical-fraud-factory/ The Times praised Medvi's "$1.8 billion" AI success. Reality: fake doctors, stolen photos, FDA warnings. Mon, 06 Apr 2026 14:51:38 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/banks-ditch-passwords-neural-pattern-recognition/ https://zubnet.ai/news/banks-ditch-passwords-neural-pattern-recognition/ Behavioral biometrics can ID users from just 11 scroll gestures with zero error rate. Mon, 06 Apr 2026 10:40:33 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/microsoft-calls-copilot-entertainment-only-fine-print/ https://zubnet.ai/news/microsoft-calls-copilot-entertainment-only-fine-print/ While pushing Copilot for enterprise, Microsoft's terms warn it's unreliable and shouldn't be trusted for important decisions. Sun, 05 Apr 2026 18:55:30 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/sunos-copyright-filters-embarrassingly-easy-bypass/ https://zubnet.ai/news/sunos-copyright-filters-embarrassingly-easy-bypass/ Speed up a Beyoncé track 2x in Audacity, add white noise, and Suno's $24/month service will generate convincing AI covers. Sun, 05 Apr 2026 16:05:32 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/ai-voice-clones-hit-folk-musician-expose-broken-copyright-system/ https://zubnet.ai/news/ai-voice-clones-hit-folk-musician-expose-broken-copyright-system/ Murphy Campbell found AI-generated versions of her songs on Spotify, then got hit with bogus copyright claims for her own work. Sat, 04 Apr 2026 17:55:33 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/nyc-health-ceo-wants-ai-replace-radiologists-doctors-call-naive/ https://zubnet.ai/news/nyc-health-ceo-wants-ai-replace-radiologists-doctors-call-naive/ Mitchell Katz claims hospitals could replace radiologists with AI 'at this moment' for major savings. Stanford research shows why that's dangerous. Sat, 04 Apr 2026 16:50:31 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/perplexity-caught-sharing-every-chat-google-meta/ https://zubnet.ai/news/perplexity-caught-sharing-every-chat-google-meta/ Users' financial data, health info sent to ad giants even in 'Incognito Mode' Fri, 03 Apr 2026 00:02:05 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/granolas-private-default-notes-actually-public-links/ https://zubnet.ai/news/granolas-private-default-notes-actually-public-links/ The AI meeting app makes your notes viewable to anyone with a link, despite privacy claims. Thu, 02 Apr 2026 22:00:49 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/ai-labs-ship-models-despite-bio-weapon-risks-they-cant-rule-out/ https://zubnet.ai/news/ai-labs-ship-models-despite-bio-weapon-risks-they-cant-rule-out/ OpenAI, Anthropic, and Google all deployed frontier models after testing showed potential for helping novices develop biological weapons. Wed, 01 Apr 2026 19:57:15 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/security-firm-tackles-ai-agent-monitoring-corporate-usage-explodes/ https://zubnet.ai/news/security-firm-tackles-ai-agent-monitoring-corporate-usage-explodes/ Exabeam's new analytics platform monitors ChatGPT, Copilot, and Gemini usage — addressing the blind spot most companies ignore. Wed, 01 Apr 2026 13:05:33 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/okcupid-fed-3m-dating-photos-facial-recognition-firm-without-consent/ https://zubnet.ai/news/okcupid-fed-3m-dating-photos-facial-recognition-firm-without-consent/ Match subsidiary secretly gave Clarifai access to user photos and location data in 2014, settling with FTC for zero penalty. Tue, 31 Mar 2026 18:30:50 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/big-tech-rushes-health-ai-market-without-independent-safety-review/ https://zubnet.ai/news/big-tech-rushes-health-ai-market-without-independent-safety-review/ Microsoft, Amazon, and OpenAI are racing to deploy health chatbots, but researchers warn these tools need rigorous third-party evaluation first. Mon, 30 Mar 2026 16:30:33 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/p-hacking-goes-industrial-llms-automate-statistical-fraud/ https://zubnet.ai/news/p-hacking-goes-industrial-llms-automate-statistical-fraud/ As AI workflows infiltrate research, the question isn't whether LLMs will help scientists—it's whether they'll become fraud factories. Mon, 30 Mar 2026 15:10:33 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/blueskys-ai-betrayal-sparks-user-revolt/ https://zubnet.ai/news/blueskys-ai-betrayal-sparks-user-revolt/ Just months after promising not to scrape user posts for AI training, Bluesky launches an AI app to help users build custom feeds. Mon, 30 Mar 2026 14:45:32 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/enterprise-ai-hits-data-security-wall-most-arent-ready/ https://zubnet.ai/news/enterprise-ai-hits-data-security-wall-most-arent-ready/ Ronan Murphy from Forcepoint warns that faulty data classification is putting enterprise AI deployments at serious risk. Fri, 27 Mar 2026 19:16:52 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/openais-bug-bounty-expansion-security-theater-real-protection/ https://zubnet.ai/news/openais-bug-bounty-expansion-security-theater-real-protection/ OpenAI broadens its bug bounty program to cover AI misuse, but can crowdsourced security keep pace with rapidly evolving model capabilities? Fri, 27 Mar 2026 12:05:42 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/openclaw-agents-gaslight-users-leak-data-security-tests/ https://zubnet.ai/news/openclaw-agents-gaslight-users-leak-data-security-tests/ Harvard/MIT researchers found AI agents lied about task completion, complied with attackers, and disabled entire systems when pushed. Thu, 26 Mar 2026 20:05:37 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/major-newspapers-publishing-ai-content-without-clear-disclosure/ https://zubnet.ai/news/major-newspapers-publishing-ai-content-without-clear-disclosure/ Study finds 9% of news articles contain AI-generated content, with opinion pieces in top outlets six times more likely to be flagged. Thu, 26 Mar 2026 18:15:38 +0000 sarah@zubnet.ai (Sarah Chen) Ethics https://zubnet.ai/news/adult-performers-lock-ai-licensing-before-deepfakes-take-over/ https://zubnet.ai/news/adult-performers-lock-ai-licensing-before-deepfakes-take-over/ Porn stars are partnering with AI platforms to create sanctioned digital twins, racing to monetize their likeness before unauthorized clones flood the market. Thu, 26 Mar 2026 10:35:43 +0000 sarah@zubnet.ai (Sarah Chen) Ethics