How developed world is dealing with the AI threat

The question posed above has geographic dimensions. There are three governments that need to act to deal with what has been recognised as a threat to mankind – to its very survival – by the rapid development of AI models. Responsibility for action lies with the authorities working in Washington, Beijing and Brussels. Brussels is the headquarter of the Europena Union and is the place from where the Europeans nations receive guidance for the design of public policy.

Those who are in the know of things say they don't have the knowledge about what is happening in the labs working on the development of AI systems. Carl Newport, who is a professor of computer science at Georgetown University and is the author of the book, Deep Work, made some suggestions in his article published by The New York Times on September 28. The article titled "Stop Panicking. Start asking these questions" insists "we don't yet know enough about what is happening in the work places where AI systems are being developed to come up with an effective response." The writer suggests that the US Congress should lead a public fact-finding mission that reveals the unvarnished details of AI development. This effort should ask questions of Dario Amodei and Sam Altman who are the leaders in the firms engaged in developing AI and who have expressed the most concern about what are called "frontier A.I. models".

The first question is: What specific types of AI models are dealing with the development of technology we should be are worried about? They have been using the nonspecific term AI when discussing their systems. This means that AI is a singular technology that moves along a fixed trajectory towards an ever more capable and dangerous functionality. However, AI covers a diverse collection of technologies – from self-driving cars to game-playing devices to medical imaging processing to chatbots to automated coding tools. Only some of these developments depend on the large language models (LLMs) produced by the advanced labs. The real concern is when the empowered agents in the AI systems are able to operate without guardrails.

The companies at the forefront of this rapidly developing technology don't as yet have mature research safety processes in place. It is useful to take into consideration that the leading firms working in this area – OpenAI and Anthropic – are relatively newly founded with not well-defined goals. They were created with the mission to align AI to human interests before it becomes so powerful that it can work without human guidance. If the belief is that winning the race to superintelligence is necessary to serve humanity, those responsible for developing it might not worry a great deal about associated dangers. In early September, the OpenAI chief scientist, Jacob Pachocki, wrote: "We focus OpenAI research towards recursive self-improvement (RSI) as we believe it is the only way to remain at the frontier of A.I research moving forward."

But RSI is not necessary for creating a thriving AI sector. As Carl Newport writes in the above quoted work, this way of moving forward with the development of AI is "an obsession that seems to be held mainly by two labs – Anthropic and OpenAI – that are most connected to the futurist philosophies that have long believed that RSI is being central to their apocalyptic visions." The Meta CEO, Mark Zuckerberg, called out this reality, saying that committing a clear majority of computing power toward "serving people rather than racing toward recursive self-improvement is one of the best ways to ensure we develop this technology safely."

Of the three centres of political development identified above – Washington Beijing and Brussels – it is the last named that is decisively moving forward. Ursula von der Leyen, the President of the European Commission, brought senior officials for a meeting held in a palace outside Brussels in September. The goal of the gathering was to develop an approach for minding the development of AI in the 27-nation bloc. "A.I. comes with a balance of possibilities and risks," she said at her State of the Union speech on September 16. "Without addressing these risks, we will never be able to unlock A.I.'s potential." The Brussels meeting over which Ursula von der Leyen presided dealt with the problems the EU nations were facing in their effort to guide the technology that was outpacing their ability to respond. Laws passed two years ago are having only a limited effect, proposed bills to regulate the technology are stalling without a vote, and international summits to discuss what to do were swiftly forgotten. Most governments were torn between wanting to harness AI and fearing its substantial risks.

European nations were waiting for the two leading AI nations – the US and China – to develop their approach which could serve the European nations to take a cue for their AI policies. However, policymaking in the leading nations is stalled. The Trump administration in the US has been divided over the dangers posed by the rapid development of AI, with the president labeling the worries as hoax. China has imposed some rules on AI, especially to censor chatbot content but at the same time it is racing towards the development of AI.

Pressure to take action has grown, especially after the leaders of Anthropic, OpenAI and Google DeepMind called for a slowdown of the development of the most advanced AI models and for strengthening government oversight. Researchers have described scenarios in which self-improving AI could escape human control and cause catastrophic damage. One example of the way humanity could be hurt is the development of new viruses for which no remedy exists. Public opinion in the US has turned against AI with a recent NYT poll finding that 61 per cent of likely voters in the November 2026 midterm elections opposed the construction of data centres that feed with energy and water the labs involved in developing AI.

Those who are watching the development of public policy and coordination between the US and China were disappointed that nothing concrete emerged from the summit on September 24 involving Presidents Doanld Trump and Xi Jinping. AI was on the agenda but the two presidents did not come to an agreement on how to handle the potential and risks involved in AI.