look news india
Amidst the blind race of Artificial Intelligence and the ongoing competition in the corporate corridors of Silicon Valley, a very sensational and serious warning has emerged from the technical world. Jacob Coxon, who has served as a top researcher at two of the world’s most prominent AI labs—OpenAI and Claude’s parent Anthropic—has warned that the development of super-intelligent AI poses an existential threat to humanity. Coxon revealed in clear words that the top scientists and developers building modern AI architectures themselves privately admit that this unbridled technology could completely destroy human civilization by the end of this decade, i.e. by the year 2030. The statement by Coxon, who recently resigned from Anthropic, a company led by Dario Amodei, has caused a stir in governments, technology regulatory bodies and the scientific community around the world.
Sharing his experience during a major international television interview, Jacob Coxon cautioned that the current process of creating superintelligent AI is akin to inviting an unnamed and unimaginable alien species to Earth. He argued that we are creating an artificial brain whose internal calculations, neural network layers and deep decision-making processes we ourselves are unable to fully understand. According to Coxon, if a scientist brings into existence a unique intelligence that is beyond human understanding and over which no human control or safety switch is possible, then this situation is frightening in itself, even without any hypothetical catastrophic scenario. He stressed that when a machine surpasses human intellectual capacity, it will take decisions based on its own priorities and safety, in which protection of human survival may be at the bottom of its list.
Jacob Coxon is not the only scientist to sound the alarm about the doomsday scenario of Artificial General Intelligence (AGI), although his direct research experience lends extraordinary weight to this warning. Elon Musk, the world’s richest entrepreneur and head of xAI (Grok), has publicly said on several forums that there is at least a 10 percent real chance of AI wiping out the human race within the next 10 years. Similarly, Geoffrey Hinton, awarded the Nobel Prize in Computer Science and known as the modern ‘Godfather of AI’, has repeatedly warned that the risk of human extinction due to uncontrolled digital superintelligence remains 10 to 20 percent. Since his resignation from Google, Hinton has been running a worldwide public awareness campaign regarding AI alignment and safety protocols.
Sensing this growing threat, Anthropic Chief Executive Officer (CEO) Dario Amodei, OpenAI CEO Sam Altman and Elon Musk have initiated discussions at the industry level to slow down the pace of AI model training and strengthen security standards. Geoffrey Hinton, along with luminaries such as Turing Award winner Yoshua Bengio, deep-learning expert Yann LeCun, LinkedIn co-founder Reid Hoffman, former Google CEO Eric Schmidt and Google DeepMind co-founder Demis Hassabis, have jointly appealed to governments to mandate stringent audits and global security rules before the release of frontier AI models. If these autonomous agents being trained on supercomputing clusters are not bound by international treaties and security boundaries in time, then this digital brain created by humans can become the biggest disaster for its creator.