AI4Future: Top AI News (27 January-2 February)

Renowned British journalist and lead columnist for The Guardian, Larry Elliott, has penned a powerful piece on the “Sputnik Moment” and the West’s potential loss in the AI arms race to China.

This week has truly become a “Sputnik Moment,” as in response to attempts to restrict China in the AI race, Chinese developers have made several super-powerful AI models available for free. These models, significantly outperforming all existing alternatives (including ChatGPT) in terms of efficiency and cost, instantly caused the US stock market to plummet by one trillion dollars, particularly affecting shares of big tech companies like Nvidia, Meta, Alphabet, and others. This includes the DeepSeek and Qwen models from Alibaba, which have also released versions that can be deployed on local hardware without sharing information with anyone. This has triggered a tsunami in the market. Key platforms aggregating leading innovations (such as Hugging Face, GitHub, ModelScope, and Azure AI Foundry) have already uploaded the Chinese models, despite rumours suggesting that the Chinese have used OpenAI’s developments. Referring to these rumours, The Guardian columnist Larry Elliott wrote: "Whatever the truth about DeepSeek, China’s tech sector is light years ahead on strategy and investment."

"The arrival of a budget Chinese chatbot called DeepSeek to rival ChatGPT really is a?wake-up call. It is a wake-up call to the US tech giants. It is a wake-up call to Wall Street. It is a wake-up call to any developed nation keen to enter the AI race.

All of that is true even if it turns out that DeepSeek is not all it is cracked up to be. If it is the real deal then the Chinese have created a premium-quality AI product that is available freely and at a tiny cost. That’s quite something. In 1957, the US was stunned when the Soviet Union became the first country to launch an artificial satellite. It has been equally taken aback by the arrival of DeepSeek. Truly this would be a Sputnik moment.

In the early days of its rapid economic development,?China?was seen as the place where US and European companies could outsource production. Labour was cheap and moving manufacturing offshore held out the promise of higher profits. The idea was that all the really advanced stuff – the product design, the research and development – would be done in the west. It would only be the assembly work that would end up in Guangdong. The creativity needed to come up with new ideas and new products would be stifled by China’s Marxist-Leninist system.

That has proved to be too complacent a view. In 2023, China?filed more patents?than the rest of the world put together. Chinese universities are turning out on average?more than 6,000 PhDs?in Stem (science, technology, engineering and maths) a month – more than double the number in the US. As DeepSeek shows, China has a growing cadre of very bright people quite capable of thinking outside the box when they are allowed to do so, as has certainly been the case with the development of AI, lithium-ion batteries and electric vehicles..." concludes Larry Elliott.

Returning to the market’s reaction, two days after the Chinese releases, OpenAI made a statement alleging that the Chinese had used ideas from ChatGPT in their DeepSeek solution, while also announcing ChatGPT Gov for US government agencies. This move was a response to the secure Chinese models, which are designed to work with confidential data and can be deployed on private hardware. OpenAI has faced criticism for lacking a similar modification.

A round-up of the key developments.

US restricts Switzerland’s access to AI chips

The US recently changed the rules for the export of products related to artificial intelligence. Only countries that are considered allies are now allowed to access these computer chips. And Switzerland is not one of them. Only 18 countries are considered trustworthy allies in the US, including France, Germany and Japan. According to Washington, these nations are worthy of unrestricted access to these very powerful computer chips, which are manufactured exclusively by US companies. This new regulation will come into force in four months. After that, Switzerland will still be able to import these chips, but will be subject to a limited quota in the coming years.

More

China's DeepSeek just dropped a free challenger to OpenAI's o1 – here's how to use it on your PC

Chinese AI startup DeepSeek this week unveiled a family of LLMs it claims not only replicates OpenAI's o1 reasoning capabilities, but challenges the American model builder's dominance in a whole host of benchmarks. Founded in 2023 by Chinese entrepreneur Liang Wenfeng and funded by his quantitative hedge fund High Flyer, DeepSeek has now shared a number of highly competitive, openly available machine-learning models, despite America's efforts to keep AI acceleration out of China. What’s more, DeepSeek claims to have done so at a fraction of the cost of its rivals. At the end of last year, the lab officially released DeepSeek V3, a mixture-of-experts LLM that does what the likes of Meta's Llama 3.1, OpenAI's GPT-4o, and Anthropic's Claude 3.5 Sonnet can do. Now it's released R1, a reasoning model fine-tuned from V3.

More

AI-based automation of jobs could increase inequality in UK, report says

Government intervention key to supporting businesses through transition, research by thinktank suggests. The automation of millions of jobs will increase inequality in the UK unless the government intervenes to support small businesses and workers through the transition, according to a report into the future of work. Ministers need to act in the interest of those who will be made unemployed or whose jobs dramatically change, says the report by the Institute for the Future of Work (IFOW) thinktank, in order to prevent skills shortages hitting employers and workers from suffering a decline in job satisfaction and wellbeing. Artificial intelligence software is expected to become a widespread tool in factories, offices and in the public sector, demanding new skills, the IFOW said. However, a survey of 5,000 UK employees found “a pervasive sense of anxiety, fear and uncertainty” about the introduction of AI technology, and what it could do to their work.

More

Another OpenAI researcher quits—claims AI labs are taking a ‘very risky gamble’ with humanity amid the race toward AGI

OpenAI safety researcher Steven Adler announced on Monday he had left OpenAI late last year after four years at the company. In a post shared on X, Adler criticized a race toward AGI that is taking shape between leading AI labs and global superpowers. “An AGI race is a very risky gamble, with huge downside,” he said. “No lab has a solution to AI alignment today. And the faster we race, the less likely that anyone finds one in time.” Alignment is the process of keeping AI working toward human goals and values, not against them.

More

Biased AI rejected my job applications, says ex-MP

A former cabinet minister who lost his seat at the general election claims he has been automatically rejected for jobs by Artificial Intelligence (AI) software because he does not have a degree. David TC Davies, who was Welsh secretary in Rishi Sunak's Conservative government, has called on employers to rethink the role of AI when recruiting. Many companies routinely use applicant tracking systems to sift and grade CVs, despite concerns they could be filtering out the best candidates. The government has produced guidelines on the use of AI in recruitment, external, which warns companies: "At all stages there is a risk of unfair bias or discrimination against applicants."

More

Alibaba releases AI model it says surpasses DeepSeek

Chinese tech company Alibaba, opens new tab on Wednesday released a new version of its Qwen 2.5 artificial intelligence model that it claimed surpassed the highly-acclaimed DeepSeek-V3. The unusual timing of the Qwen 2.5-Max's release, on the first day of the Lunar New Year when most Chinese people are off work and with their families, points to the pressure Chinese AI startup DeepSeek's meteoric rise in the past three weeks has placed on not just overseas rivals, but also its domestic competition. "Qwen 2.5-Max outperforms ... almost across the board GPT-4o, DeepSeek-V3 and Llama-3.1-405B," Alibaba's cloud unit said in an announcement posted on its official WeChat account, referring to OpenAI and Meta's most advanced open-source AI models.

More

Introducing ChatGPT Gov: designed to streamline government agencies’ access to OpenAI’s frontier models

Today we’re announcing ChatGPT Gov, a new tailored version of ChatGPT designed to provide U.S. government agencies with an additional way to access OpenAI’s frontier models. Agencies can deploy ChatGPT Gov in their own Microsoft Azure commercial cloud or Azure Government cloud on top of Microsoft’s Azure’s OpenAI (opens in a new window) Service. Self-hosting ChatGPT Gov enables agencies to more easily manage their own security, privacy, and compliance requirements, such as stringent cybersecurity frameworks (IL5, CJIS, ITAR, FedRAMP High). Additionally, we believe this infrastructure will expedite internal authorization of OpenAI’s tools for the handling of non-public sensitive data.

More

The west is already losing the AI arms race. Whatever the truth about DeepSeek, China’s tech sector is light years ahead on strategy and investment

All of that is true even if it turns out that DeepSeek is not all it is cracked up to be. If it is the real deal then the Chinese have created a premium-quality AI product that is available freely and at a tiny cost. That’s quite something. In 1957, the US was stunned when the Soviet Union became the first country to launch an artificial satellite. It has been equally taken aback by the arrival of DeepSeek. Truly this would be a Sputnik moment.

More

DeepSeek R1 is now available on Azure AI Foundry and GitHub

DeepSeek R1 is now available in the model catalog on Azure AI Foundry and GitHub, joining a diverse portfolio of over 1,800 models, including frontier, open-source, industry-specific, and task-based AI models. As part of Azure AI Foundry, DeepSeek R1 is accessible on a trusted, scalable, and enterprise-ready platform, enabling businesses to seamlessly integrate advanced AI while meeting SLAs, security, and responsible AI commitments—all backed by Microsoft’s reliability and innovation.

More

Meta announces Mesa Data Center is operational

Meta announced on Jan. 30 that its Mesa Data Center is now serving traffic. That means that this data center is now part of the company’s global infrastructure that brings its technologies and services to life, making it possible to connect billions of people worldwide. Meta has been a part of the Mesa community since breaking ground on the data center in 2021. “We came to Mesa four years ago because it offered excellent infrastructure, access to renewable energy, a strong pool of talent for both construction and operations staff, and great community partners that have helped us move forward quickly, and we’re thrilled to call it home,” said Brad Davis, data center community and economic development director at Meta.

More

China’s ‘Artificial Sun’ Sets Record by Sustaining 100-Million-Degree Heat for 18 Minutes

Chinese scientists achieved a breakthrough by maintaining plasma at 100 million degrees Celsius for 18 minutes, advancing the quest for limitless clean energy. Chinese researchers have made a significant leap in nuclear fusion technology by sustaining plasma temperatures of over 100 million degrees Celsius for nearly 18 minutes. This groundbreaking achievement was made possible through the Experimental Advanced Superconducting Tokamak (EAST), a reactor often referred to as the “artificial sun.” The experiment, conducted at the EAST facility, mimicked the Sun’s fusion process using hydrogen and deuterium gases as fuel. By fusing light atomic nuclei at extreme temperatures, the reactor aims to replicate the energy generation of stars. Song Yuntao, Director of EAST, explained the importance of this achievement: “A fusion device must achieve stable operation at high efficiency for thousands of seconds to enable the self-sustaining circulation of plasma, which is critical for the continuous power generation of future fusion plants.”

More