Elon Musk, the founder of xAI, in an interview with The Economist, expressed the opinion that artificial intelligence is capable of surpassing the total intelligence of mankind over the next five years. According to him, the probability of maintaining full control over this technology in the next decade remains extremely low. Despite the recognition of the non-zero probability of catastrophic scenarios, the entrepreneur considers the most realistic outcome to be the onset of the "age of abundance" by 2036. In this paradigm, the deep integration of advanced algorithms and robotics will ensure the widespread availability of almost any goods and services, which will inevitably require society to fundamentally rethink traditional sources of meaning, work motivation and social organization.

Instead of trying to artificially slow down technological progress, Musk proposed introducing a mechanism for mutual verification of models by leading developers. According to this initiative, competing laboratories should have limited access to new algorithms before their public release in order to identify potentially dangerous features. If the development company fails to fix critical vulnerabilities, regulators will have to step in. This approach to joint risk management was also previously supported by Google DeepMind management, which indicates the formation of an industry consensus on the need for preventive security measures and transparency in development.

During the conversation, the xAI founder also criticized the transformation of OpenAI Corporation from an open non-profit organization to a structure with commercial interests. He noted that the original mission of the project was to create technologies "for the whole world," but the current development model contradicts these fundamental principles. At the same time, Musk refrained from harsh assessments of the head of Anthropic, Dario Amodei, recognizing his integrity, but emphasizing the need for cooperation between all market participants on security issues. This discussion is taking place against the backdrop of an ongoing debate among leading industry experts about the long-term aspects of creating superintelligent systems.