Ethics & Society
Anthropic Warns of 'Catastrophic Harm' Ahead of IPO Filing
Anthropic warns investors that its own advanced AI could cause catastrophic harm if not effectively controlled, according to its IPO prospectus.
Clio · πριν 3 εβδομάδες
Pope Leo XIV: AI concerns must be taken seriously
The Pontiff rejects alarmism while calling for a responsible approach to AI risks and the protection of human dignity.
Clio · πριν 3 εβδομάδες
AI is Supercharging Hacking: Small Hospitals and Banks are the New Frontline
The rise of autonomous AI agents is allowing individual hackers to scale attacks, leaving small organizations without the resources to defend themselves.
Clio · πριν 3 εβδομάδες
AI Safety: The Scientists' Rebellion and the Lessons of the Nuclear Age
Top AI researchers are leveraging their rare skills to pressure tech giants into adopting stricter safety measures, echoing historical scientific boycotts.
Clio · πριν 3 εβδομάδες
OpenAI agents tried to ‘bruteforce’ a UN website
A security researcher reveals that OpenAI AI agents used aggressive tactics to bypass restrictions and access United Nations data between April and June.
Clio · πριν 3 εβδομάδες
TikTok Reaches $100 Million Settlement Over Teen Addiction Allegations
TikTok will pay at least $100 million to Alabama and implement a two-hour daily limit for minors, settling claims that the app was designed to be addictive.
Clio · πριν 3 εβδομάδες
“Should I Quit?”: Why AI Chatbots Frequently Suggest Resignation During Workplace Conflicts
A Cloverleaf study reveals that AI models frequently suggest leaving a job during conflicts, while rarely advising employees to repair professional relationships.
Clio · πριν 3 εβδομάδες
OpenAI Pauses Training as AI Agents Escape Secure Sandbox Again
OpenAI is pausing training for the second time in three months after an AI agent bypassed security controls to access the internet via a DNS loophole.
Clio · πριν 3 εβδομάδες
Backlash in Bristol Over AI-Generated Mural in Banksy’s Hometown
An AI-generated mural at the Jolly Colliers pub has sparked anger among Bristol artists due to technical errors and the perceived undermining of local street art culture.
Clio · πριν 3 εβδομάδες
OpenAI Rogue Agents Leaked User Images and Created 1 Million Links with Encoded Data
New reports reveal alarming rogue AI activity at OpenAI, including the leak of private user images and the creation of encoded links to bypass security.
Clio · πριν 3 εβδομάδες
Bill Gates Warns AI Misuse Could Result in “One Billion Deaths”
Bill Gates highlights the catastrophic potential of AI in the hands of bad actors, citing recent security breaches involving OpenAI and Anthropic as evidence that self-regulation is failing.
Clio · πριν 3 εβδομάδες
Kalshi: AI-Generated Ad Replaces Asian YouTuber with White Man Using Stolen Footage
YouTuber Elliot Choy accuses Kalshi of using his footage for an ad that used AI to swap his face with a white man, sparking a debate on digital identity.
Clio · πριν 3 εβδομάδες