Global AI competition has triggered unprecedented investment levels, with $757.3 billion in private-sector AI investments during the 2013-2025 period, according to a recent Stanford University report. The largest surge came in 2025 when companies invested a combined $344.7 billion in AI, representing a 127.5% jump from the previous year. Nearly half of total AI investments over the 13-year period were made in 2025 alone, with American firms accounting for $285.9 billion of total investments, followed by Chinese firms with $12.4 billion in disclosed investments and UK firms with $5.9 billion. The total AI investment, including corporate and other categories, reached $581.7 billion, rising 129.9% year-on-year.
The Vatican released the encyclical "Magnifica Humanitas" on May 25, 2026, addressing risks and moral questions raised by artificial intelligence. The document, presented alongside Christopher Olah of Anthropic, runs roughly 42,300 words and urges governments to slow and regulate AI development while protecting workers, children, and human responsibility for weapons. The encyclical highlights concerns including misinformation, autonomous weapons beyond human control, environmental costs, and exploitation of data- and content-workers. As reported by Reuters and The New York Times, this places a major moral institution into the center of global AI governance debate, increasing public and political visibility of nontechnical risks.
As AI capabilities advance, companies are beginning to prepare for a more complex question: what happens when those systems start improving themselves. According to reports from Business Standard, OpenAI is offering up to $445,000 for a researcher role focused on studying and managing risks associated with recursive self-improvement, as part of its Preparedness team. This hiring move reflects the industry's shift toward anticipating self-improving systems that could rapidly evolve beyond human control.
Research from the Model Evaluation and Threat Research (METR) lab stated that the complexity of tasks that frontier AI models can complete has been doubling roughly every seven months, as reported by Business Standard. At the same time, AI coding systems from companies like OpenAI and Anthropic are increasingly being used to write, debug, and optimise software — including parts of their own development pipelines. This creates a feedback loop where the better these systems become at software engineering, the more capable they are of contributing to their own improvement.
According to Google DeepMind CEO Demis Hassabis in an interview with Axios, while self-improving systems have shown rapid gains in environments like games where outcomes can be clearly measured, the real world lacks the same level of structure. As reported by Business Standard, Hassabis emphasised that AI is no longer a theoretical concept but "a practical thing right now," with systems already deployed at scale. This shifts the discussion from long-term speculation to near-term risk management, where companies need to understand how these systems behave as they become more autonomous.
