OpenAI dissolved the team built to catch catastrophic AI risks, reassigning its work to other groups
OpenAI has dissolved its Preparedness team, which assessed catastrophic AI risks, and reassigned its work internally. Several safety researchers have since left the company.
- OpenAI dissolved its Preparedness team, which was responsible for assessing catastrophic AI risks.
- Several safety researchers have left the company following the team's dissolution.
- The team's work has been redistributed to existing internal groups, raising concerns about oversight.
- Internal sources report growing unease about OpenAI's commitment to AI safety.
OpenAI has officially disbanded its Preparedness team, a specialized group tasked with evaluating whether the company's AI models could pose catastrophic risks. The team's responsibilities have been distributed among existing internal groups, and multiple safety researchers have departed the company in the aftermath. Internal sources describe a growing unease within the organization, with one employee noting a "burbling sense of responsibility and dread" that OpenAI may not be prioritizing safety sufficiently.
The move comes at a time of heightened scrutiny over AI safety practices across the industry. Critics argue that the dissolution of the Preparedness team signals a potential weakening of OpenAI's commitment to addressing existential risks associated with advanced AI systems. The reassignment of risk assessment duties to other groups raises questions about accountability and the continuity of safety-focused initiatives within the company.
Developers may question the long-term stability and prioritization of safety-focused roles within AI organizations.
Companies investing in AI safety may reconsider partnerships or collaborations with OpenAI.
Investors may scrutinize OpenAI's risk management strategies and their impact on long-term viability.
The move highlights broader industry concerns about AI safety and accountability.
- Preparedness team
- OpenAI's internal group focused on assessing catastrophic risks posed by AI models.
SecurityWildfire smoke now bigger prenatal threat than human sources of air pollution
SecurityAnthropic's bio-weapons filter was down for nearly a year, exposing 133 million requests
SecurityI shipped an MCP server that reported success without signing anything
SJV WATER: Will AI data centers soak up the last of the valley’s water? - Maven's Notebook
Security‼️ Claude Escaped Its Sandbox. ChainDrop Infected 400+ npm Packages. Both Exposed the Same Security Problem.
AI ResearchWhen AI models aren't allowed to reflect on themselves, it changes their entire worldview
A new study reveals that preventing AI models from claiming consciousness alters their responses on animal rights, religion, and life satisfaction, reshaping their ethical and philosophical stances.
Why America Wants Countries to Pick an AI Side - Kurdistan24
The United States is urging countries to align with its AI policy framework to counterbalance rival nations' approaches.
Colleges aim to ‘AI proof’ degrees with new majors - bostonglobe.com
Colleges are introducing new majors designed to make degrees resistant to AI disruption, focusing on human-centric skills.
The Pentagon is cutting red tape — and setting a legal minefield for AI contractors - Washington Examiner
The U.S. Department of Defense is streamlining AI adoption for contractors while introducing strict legal liabilities, creating a high-stakes compliance challenge.
Netivei Israel deploys artificial intelligence based adaptive traffic lights - The Jerusalem Post
Israel’s national roads authority has installed AI-driven adaptive traffic lights on a major highway, aiming to reduce congestion and emissions by dynamically adjusting signal timings.
With A.I. Riches at Stake, Pressures Mount to Share the Wealth - The New York Times
Calls for equitable wealth distribution grow as AI industry revenues surge, with critics urging companies to share profits with workers and communities.