The Guardian view on a global AI race: geopolitics, innovation and the rise of chaos | Editorial
eYears ago, Vladimir Putin declared that mastering artificial intelligence (AI) would make a nation “Ruler of the worldWestern technology sanctions after Russia’s invasion of Ukraine should have stopped his ambitions to drive artificial intelligence 2030. But this may be too hasty to judge. Last week, Chinese lab Deepseek unveiled R1, an AI Analysts Competitors say Openai’s best thinking model, O1. Surprisingly, it matches the capabilities of the O1 while using a fraction of the computing power — and a tenth of the cost. As expected, one of the first is Putin Moves In 2025, there was agreement with China on the development of artificial intelligence. It seems that the launch of R1 is no coincidence, as Donald Trump gave $500 billion from Donald Trump Stargate She plans to outperform her peers. Openai has customized Deepseek’s parent, High Capital Bulletinas a potential threat. But at least three Chinese The labs claim to rival or exceed Openai’s achievements.
Expect tighter US chip sanctions, Chinese companies stock Processors are critical to ensuring that AI models may advance despite access to hardware. Deepseek’s success underscores creativity born of necessity: Lacking big data brands or powerful specialized chips, it has made breakthroughs by activating better data and improving its model. Unlike proprietary systems, the source code for R1 is public, allowing any qualified person to modify it. However, its openness has limits: Admin By China’s Internet regulator, R1 conforms to “core socialist values.” Type in Tiananmen Square or Taiwan, and the form is said to close the conversation.
Deepseek’s R1 highlights a broader discussion about the future of AI Ownership walls, controlled by a few large companies, or be “Open sourced“To promote global innovation? One of Biden The administration’s final actions were to phase out open source AI on national security grounds. Artificial intelligence can be freely accessed, and bad actors can be empowered. Interestingly, Mr. Trump later rescinded the order, arguing that open source development planning hurts innovation. Open source advocates, Like deadyou have a point when crediting recent AI breakthroughs to a decade of freely sharing code. However, the risks are undeniable: In February, Openai shut down accounts linked to state-backed hackers from China, Iran, Russia, and North Korea that used its tools in phishing and malware campaigns. By summer, Openai to stop services in those countries.
America’s superior control over critical AI hardware in the future may give its rivals little chance to compete. Openai Offers”Organized access“, controlling how users interact with its models. But Deepseek’s success suggests that open-source AI could lead the way innovation Through creativity, rather than brute processing power. The contradiction is clear: open source democratizing intelligence gives technology and fuel, but it also enables exploitation by Malefactors. Resolving this tension between innovation and security requires International framework To prevent misuse.
The AI race is as much about global influence as technological dominance. Mr. Putin urges developing countries to unite challenge US technical leadership, but without global regulation, there are enormous risks in a frenzied push for AI supremacy. It would be wise to pay Geoffrey Hintona pioneering AI and Nobel laureate. Warns of the pace of collapse progress Reduces the potential for disaster. In the race to control this technology, the biggest risk is not far behind. She completely loses control.
