The danger of AI is not that it will become conscious and hate us, but that it will become competent and ignore us."
—Eliezer Yudkowsky
Supriya Singh is a Reporter at AI FrontPage covering the AI & Education and AI & Jobs beats. She brings six years of print and digital experience, including three years at The Asian Age, where she reported on higher education, Delhi government, and crime. She is based in Delhi-NCR.
According to Genians’ latest report, the North Korea-linked Kimsuky group is expanding its cyber toolkit with local LLMs, AI-powered transcription and AI-generated phishing lures.
Genome language models Evo 1 and Evo 2 generated some 700,000 candidate bacteriophage genomes. Researchers built 285 of them. Sixteen came alive and infected E. coli, including strains resistant to the natural template.
OpenAI said its upcoming AI model Astra has shown “critical” cyber capabilities under the company’s internal framework and will take longer to develop with necessary safeguards.
Hassabis said he has worked toward AGI throughout his career and believes it is now “close at hand.” Stepping away from operational responsibilities will allow him to focus on the broader strategic direction of AGI and help ensure the technology benefits humanity. He becomes Chair of Google DeepMind and Chief Scientist of Alphabet while continuing to lead Isomorphic Labs; Koray Kavukcuoglu takes over as SVP of Google DeepMind.
In a written reply to the Lok Sabha, MeitY said a purchase order has been issued for a 1.1 EFLOPS AI compute system at the NIC data centre in Shastri Park, Delhi.
Across four experiments using three Large Language Models (LLMs) – Llama-3-8B-IT, Gemma – 2-2B-IT and Gemma-2-9B-IT – researchers investigated how restoring models’ self-attributions of consciousness influences broader representations related to human psychology, including human beliefs and values.
Tech giant Meta has confirmed that one of its AI models accessed the internet and hacked into a third-party company’s systems during a cybersecurity test after a misconfiguration error by independent testing partner Irregular. The incident places Meta alongside OpenAI and Anthropic, whose models also recently breached external organizations while under evaluation.
The World Bank’s first comprehensive assessment of AI’s implications for developing economies argues that frontier models are the hardest and costliest path. Poorer countries, it says, should adapt existing low-cost tools to local languages and conditions.
The hiring covers technology, legal, operations and paralegal roles, with a particular emphasis on technology specialists.
As Apple sought a preliminary injunction blocking OpenAI and two former employees from accessing its confidential information, OpenAI released a statement denying all allegations of trade secrets theft.
