Cybersecurity researchers have discovered a new AI jailbreak technique called 'Bad Likert Judge' that can boost the success rates of malicious prompts bypassing large language models' safety guardrails by over 60%, enabling the generation of potentially harmful or illegal content.
DeepMind's CEO Hassabis predicts human-level artificial general intelligence (AGI) within a decade, heralding a 'golden era' where AGI solves major global challenges like disease and energy scarcity, though potential risks remain.
The CEO of AI firm Anthropic, Dario Amodei, warns that AI could eliminate 50% of entry-level office jobs soon, part of a Silicon Valley script hyping AI's potential to fix everything after ruining everything first, despite lack of evidence to support such dramatic claims.
Anthropic CEO Dario Amodei cautions that the rapid advancement of AI technology poses a significant threat to mass employment, urging society to address the potential for AI to automate numerous tasks and displace human workers across various industries.
Netflix co-founder Reed Hastings joins Anthropic's board of directors, bringing his extensive experience in scaling Netflix globally and serving on tech giants' boards to help guide the AI company's approach to beneficial AI development.
Protesters in San Francisco warn that the increasing intelligence of artificial intelligence (AI) poses an existential threat to humanity, calling for regulation, while President Trump's 'One Big Beautiful Bill Act' threatens to override state efforts to regulate AI for a decade.
Anthropic's Claude Opus 4 AI model sets a new record by coding nonstop for seven hours, outperforming OpenAI's GPT-4.1 on a software engineering benchmark, reshaping enterprise AI with its sustained focus and reasoning capabilities.
While AI risks deserve scrutiny, media coverage often sensationalizes dystopian scenarios, influenced by billionaire-backed movements, neglecting nuance; more balanced, evidence-based reporting highlighting diverse expert voices is needed to improve public understanding.
House Republicans pushed a controversial bill imposing a decade-long freeze on state AI regulations, sparking concerns over potential erosion of consumer safeguards and states' ability to address AI risks like deepfakes and algorithmic bias.
Anthropic launches a bug bounty program offering up to $25,000 in rewards to researchers who can test and expose vulnerabilities in its AI safety measures designed to prevent the misuse of its models for CBRN (chemical, biological, radiological, and nuclear) weapons development.