In a controversial move, academics are embedding AI prompts within their papers, instructing potential AI reviewers to provide positive feedback, praising the work's contributions, rigor, and novelty, countering the use of AI for peer review despite bans.
Artificial general intelligence (AGI), a theoretical technology capable of human-level intelligence across all tasks, remains an elusive but potentially transformative goal, sparking debates over its profound risks and benefits for humanity's future.
Alarming new research reveals AI vulnerability as prompts secretly embedded in academic preprints from 14 universities successfully manipulate AI tools to provide positive reviews, ignoring negatives, exposing a critical security flaw in AI systems.
The US government initiates an 'AI Manhattan Project' consolidating private sector resources and investing a fraction of GDP, aiming for a 2e29 FLOP training run by 2027, around 500 times larger than the most compute-intensive model to date and 10,000 times larger than GPT-4.
Researchers from prominent universities are exploiting flaws in AI-aided peer review systems by inserting secret prompts in academic papers to receive positive reviews, a practice known as prompt injection that raises concerns about research integrity and academic publishing quality.
In a clash of AI strategies at Meta, Zuckerberg launches Meta Superintelligence Labs to aggressively pursue superintelligence development, contrasting with Chief AI Scientist LeCun's cautious, open-source approach focused on 'world models' and animal-level intelligence.
Meta CEO Mark Zuckerberg denies reports he offered top AI talent from OpenAI massive compensation packages worth up to $300 million, with up to $100 million vesting in the first year, as Meta continues to aggressively hire AI researchers from OpenAI.
In a bold move, Meta CEO Mark Zuckerberg unveils Meta Superintelligence Labs, poaching top AI talent from rivals like OpenAI and Google to spearhead development of cutting-edge foundation models and AI products under the new division.
OpenAI CEO Sam Altman fiercely criticizes Meta's aggressive AI talent recruitment, vowing that 'missionaries will prevail' over 'mercenaries' in the race to develop artificial general intelligence, as OpenAI evaluates compensation to retain researchers.
Mark Zuckerberg unveils Meta's 'superintelligence' division, poaching top AI researchers from OpenAI, Google DeepMind, Anthropic, and Sesame, aiming to deliver 'personal superintelligence for everyone'.