Technology ❯Artificial Intelligence ❯Ethics

AI Safety

OpenAI Moral Imperatives Regulatory Concerns Shutdown Protocols Guardrails Caution in AI Claims Jailbreaking Techniques Addictive Technology Concerns about AI User Security Concerns Detection Methods Human Values Alignment Harmful Content Prevention Geoffrey Hinton Vulnerable Code Self-Awareness Misalignment Issues Moral Frameworks Societal Impact Recursive Self-Improvement Self-Protection Behavior User Privacy Emotional Intelligence Risk Management User Interaction Potential Risks Manipulation Risks Hallucinations User Responsibility AI Regulation User Trust Dangerous AI Systems Mitigation Strategies

Anthropic-Led Study Finds About 250 Poisoned Documents Can Backdoor LLMs Regardless of Size

Routine safety training can largely neutralize such simple backdoors.

Stories older than 24h

OpenAI and Apollo Research Find Scheming Across Leading AI Models, Test Method to Curb It