Google, OpenAI's AI models claim gold at global math competition
The unit of Alphabet's Google and OpenAI have both said their artificial intelligence models won gold medals at a global mathematics competition, signaling a breakthrough in math capabilities in the race to build systems that can rival human intelligence.
The results marked the first time that AI systems crossed the gold-medal scoring threshold at the International Mathematical Olympiad (IMO) for high-school students.
Both companies' models solved five out of six problems, achieving the result using general-purpose 'reasoning' models that processed mathematical concepts using natural language, in contrast to the previous approaches used by AI firms.While Google DeepMind worked with the IMO to have their models graded and certified by the committee, OpenAI did not officially enter the competition.
The startup revealed their models have achieved a gold medal-worthy score on this year's questions on Saturday, citing grades by three external IMO medalists.
The achievement suggests AI is less than a year away from being used by mathematicians to crack unsolved research problems at the frontier of the field, according to Junehyuk Jung, a math professor at Brown University and visiting researcher in Google's DeepMind AI unit.
'I think the moment we can solve hard reasoning problems in natural language will enable the potential for collaboration between AI and mathematicians,' Jung told Reuters.
OpenAI's breakthrough was achieved with a new experimental model centered on massively scaling up 'test-time compute.' This was done by both allowing the model to 'think' for longer periods and deploying parallel computing power to run numerous lines of reasoning simultaneously, according to Noam Brown, researcher at OpenAI. Brown declined to say how much in computing power it cost OpenAI, but called it 'very expensive.' To OpenAI researchers, it is another clear sign that AI models can command extensive reasoning capabilities that could expand into other areas beyond math.
The optimism is shared by Google researchers, who believe AI models' capabilities can apply to research quandaries in other fields such as physics, said Jung, who won an IMO gold medal as a student in 2003.Of the 630 students participating in the 66th IMO on the Sunshine Coast in Queensland, Australia, 67 contestants, or about 11%, achieved gold-medal scores.
Google's DeepMind AI unit last year achieved a silver medal score using AI systems specialized for math. This year, Google used a general-purpose model called Gemini Deep Think, a version of which was previously unveiled at its annual developer conference in May.Unlike previous AI attempts that relied on formal languages and lengthy computation, Google's approach this year operated entirely in natural language and solved the problems within the official 4.5-hour time limit, the company said in a blog post.
OpenAI, which has its own set of reasoning models, similarly built an experimental version for the competition, according to a post by researcher Alexander Wei on social media platform X. He noted that the company does not plan to release anything with this level of math capability for several months.
This year marked the first time the competition coordinated officially with some AI developers, who have for years used prominent math competitions like IMO to test model capabilities.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Qatar Tribune
11 hours ago
- Qatar Tribune
Urgent need for global approach on AI regulation: UN tech chief
Agencies The world urgently needs to find a global approach on regulating artificial intelligence, the United Nations' top tech chief said, warning that fragmentation could deepen risks and inequalities. Doreen Bogdan-Martin, head of the UN's International Telecommunications Union (ITU) agency, told AFP she hoped that AI 'can actually benefit humanity'. But as concerns mount over the risks posed by the fast-moving technology -- including fears of mass job losses, the spread of deepfakes and disinformation, and society's fabric fraying -- she insisted that regulation was key. 'There's an urgency to try to get... the right framework in place,' she said, stressing the need for 'a global approach'. Her comments came after U.S. President Donald Trump last week unveiled an aggressive, low-regulation strategy aimed at ensuring the United States stays ahead of China on more than 90 proposals, Trump's plan calls for sweeping deregulation, with the administration promising to 'remove red tape and onerous regulation' that could hinder private sector AI development. Asked if she had concerns about an approach that urges less, not more, regulation of AI technologies, Bogdan-Martin refrained from commenting, saying she was 'still trying to digest' the U.S. plan.'I think there are different approaches,' she said.'We have the EU approach. We have the Chinese approach. Now we're seeing the U.S. approach. I think what's needed is for those approaches to dialogue,' she said. At the same time, she highlighted that '85 percent of countries don't yet have AI policies or strategies'.A consistent theme among those strategies that do exist is the focus on innovation, capacity building and infrastructure investments, Bogdan-Martin said. 'But where I think the debate still needs to happen at a global level is trying to figure out how much regulation, how little regulation, is needed,' she said. Bogdan-Martin, who grew up in New Jersey and has spent most of her more than three-decade career at the ITU, insisted the Geneva-based telecoms agency that sets standards for new technologies was well-placed to help facilitate much-needed dialogue on the issue.'The need for a global approach I think is critical,' she said, cautioning that 'fragmented approaches will not help serve and reach all'. As countries and companies sprint to cement their dominance in the booming sector, there are concerns that precautions could be thrown to the wind -- and that those who lose the race or do not have the capacity to participate will be left behind. The ITU chief hailed 'mind-blowing' advances within artificial intelligence, with the potential to improve everything from education to agriculture to health care -- but insisted the benefits must be shared. Without a concerted effort, there is a risk that AI will end up standing for 'advancing inequalities', she warned, cautioning against deepening an already dire digital divide worldwide.
Qatar Tribune
4 days ago
- Qatar Tribune
Trump unveils AI plan that aims to clamp down on regulations and ‘bias'
Agencies Washington The Trump administration has unveiled a sweeping roadmap to develop artificial intelligence, pledging to boost US innovation while stripping away what it calls 'bureaucratic red tape' and 'ideological bias'. The 28-page AI Action Plan outlines more than 90 policy actions for the rapidly developing technology that administration officials say can be implemented over the next year. 'We believe we're in an AI race, and we want the United States to win that race,' Trump administration crypto czar David Sacks told reporters. The AI plan promises to build data centre infrastructure, and promote American technology - but was panned by critics who consider it an ideological flex by the White House. The plan also calls for federal agencies to review and repeal policies that stand in the way of AI development, and encourage AI in both government and the private sector. President Donald Trump is expected to sign three related executive orders on Wednesday. One order will promote the international export of US-developed AI technologies, while another aims to root out what the administration describes as 'woke' or ideologically biased AI systems. 'American development of AI systems must be free from ideological bias or engineered social agendas,' the White House said. 'With the right government policies, the United States can solidify its position as the leader in AI and secure a brighter future for all Americans.' Crypto czar Sacks added that the plan is partially focused on preventing AI technology from being 'misused or stolen by malicious actors' and will 'monitor for emerging and unforeseen risks from AI'. The Trump administration has positioned the expansion of AI infrastructure and investments in the United States as a way to stay ahead of China. 'AI is a revolutionary technology that's going to have profound ramifications for both the economy and national security,' Sacks said. 'It's just very important that America continues to be the dominant power in AI.' But critics argued that the plan was a giveaway to Big Tech. 'The White House AI Action plan was written by and for tech billionaires, and will not serve the interests of the broader public,' said Sarah Myers West, co-executive director of the AI Now Institute. '[T]he administration's stance prioritizes corporate interests over the needs of everyday people who are all already being affected by AI,' West added.
Al Jazeera
5 days ago
- Al Jazeera
Trump administration unveils wide ranging AI action plan
The administration of United States President Donald Trump has unveiled its new artificial intelligence action plan, which includes a strategy it says will boost the US standing in AI as it competes with China for dominance in the rapidly growing sector. The White House released the 25-page 'America's AI Action Plan' on Wednesday. It includes 90 different policy proposals that the administration says will increase AI tools for allies around the globe. It will also promote production of new data centres around the US. It will scrap federal regulations that 'hinder AI development', although it is not clear which regulations are in question. In a statement, US Secretary of State Marco Rubio said the plan will 'ensure America sets the technological gold standard worldwide, and that the world continues to run on American technology'. The president is expected to announce a series of executive orders which will outline key parts of the plan around 5pm in New York (21:00 GMT). 'We believe we're in an AI race … and we want the United States to win that race,' White House AI czar David Sacks told reporters on Wednesday. The White House says the plan will 'counter Chinese influence in international governance bodies' and also will give the US more control over exports of AI technology. However, the administration did not offer any details on how it plans to do that. The plan outlined by the Trump administration will also include a framework to analyse models built by China to assess 'alignment with Chinese Communist Party talking points and censorship'. Free speech in the spotlight The plan says that it will also uphold free speech in models that will allow systems to be 'objective and free from top-down ideological bias' for organisations wanting to do business with the federal government. A senior White House official said the main target was AI models that consider diversity and inclusion, according to The Wall Street Journal, which, experts say, signals the concern is the government's perceived liberal bias as opposed to an overall bias. 'The government should not be acting as a Ministry of AI Truth or insisting that AI models hew to its preferred interpretation of reality,' Samir Jain, vice president of policy at the Center for Democracy & Technology, said in a statement provided to Al Jazeera. 'The plan is highly unbalanced, focusing too much on promoting the technology while largely failing to address the ways in which it could potentially harm people.' Conservatives have long accused AI chatbots of having a liberal bias, comparable to their comments on legacy media for providing critical coverage of the administration. However, it comes as users of GrokAI, former Trump ally and right-wing tycoon Elon Musk's AI platform, have accused it of having a right-wing lean. Musk's X AI is part of a $200m package with the Pentagon that has other AI companies, including OpenAI. Building out data centres A key focus of the new plan will be to build out new data centers for AI technology as the industry rapidly expands. The administration said that will include streamlining permits for new centre development and the energy production facilities used to power these data centres. The plan sidesteps environmental concerns that have been a major criticism of the AI industry. AI 'challenges America to build vastly greater energy generation than we have today', the plan said. AI data centres have been tied to increased power consumption and, in turn, greenhouse gas emissions. According Google's 2024 sustainability report, there was a 48 percent increase in power greenhouse gas emissions since 2019 which, it says, will only become more prevalent. 'This result was primarily due to increases in data center energy consumption and supply chain emissions. As we further integrate AI into our products, reducing emissions may be challenging due to increasing energy demands from the greater intensity of AI compute, and the emissions associated with the expected increases in our technical infrastructure investment,' the report said. The streamlining of permits also comes as the US Environmental Protection Agency (EPA) plans to reverse its scientific determination that greenhouse gas emissions endanger public health. That change would remove the legal framework that climate regulations are based on, the Reuters news agency has reported, citing two unnamed sources. The reversal would remove the 'endangerment finding', making it easier for the EPA to undo legislation limiting greenhouse gas emissions on energy-producing facilities, including those used to power AI data centres. The administration has created environmental review exceptions for data centre construction and will allow expanding access to federal lands for AI development. 'AI will improve the lives of Americans by complementing their work — not replacing it,' the plan says. It, however, comes as employers across the country scrap jobs because of AI. Earlier this month, Recruit Holdings, the parent company of Indeed and Glassdoor, cut 1300 jobs which it directly attributed to AI. In June, Salesforce CEO Marc Benioff said that AI is doing 30 to 50 percent of the company's workload. In February, the tech giant laid off 1,000 employees. Analysts say the plan looks promising for investors in the AI sector. 'This is a watershed moment in the AI revolution, and Trump recognises this AI arms race between the US and China. A big step forward,' Dan Ives, analyst at Wedbush Securities, told Al Jazeera. As of 4pm in New York (20:00 GMT), stocks of AI-focused companies had mixed results. NVIDIA was up 2.1 percent; Palantir up 3.6 percent, Oracle up 1.5 percent and Microsoft was up 0.3 percent. On the other hand, Google's parent company Alphabet was down 0.5 percent.
