Humans beat AI gold-level score at top math contest
Neither model scored full marks—unlike five young people at the International Mathematical Olympiad (IMO), a prestigious annual competition where participants must be under 20 years old.
Google said Monday that an advanced version of its Gemini chatbot had solved five out of the six math problems set at the IMO, held in Australia's Queensland this month.
"We can confirm that Google DeepMind has reached the much-desired milestone, earning 35 out of a possible 42 points—a gold medal score," the US tech giant cited IMO president Gregor Dolinar as saying.
"Their solutions were astonishing in many respects. IMO graders found them to be clear, precise and most of them easy to follow."
Around 10 percent of human contestants won gold-level medals, and five received perfect scores of 42 points.
US ChatGPT maker OpenAI said that its experimental reasoning model had scored a gold-level 35 points on the test.
The result "achieved a longstanding grand challenge in AI" at "the world's most prestigious math competition," OpenAI researcher Alexander Wei wrote on social media.
"We evaluated our models on the 2025 IMO problems under the same rules as human contestants," he said.
"For each problem, three former IMO medalists independently graded the model's submitted proof."
Google achieved a silver-medal score at last year's IMO in the British city of Bath, solving four of the six problems.
That took two to three days of computation—far longer than this year, when its Gemini model solved the problems within the 4.5-hour time limit, it said.
The IMO said tech companies had "privately tested closed-source AI models on this year's problems," the same ones faced by 641 competing students from 112 countries.
"It is very exciting to see progress in the mathematical capabilities of AI models," said IMO president Dolinar.
Contest organizers could not verify how much computing power had been used by the AI models or whether there had been human involvement, he cautioned. — Agence France-Presse
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
GMA Network
7 hours ago
- GMA Network
Microsoft server hack has now hit 400 victims, researchers say
WASHINGTON - A sweeping cyber-espionage campaign organization centered on vulnerable versions of Microsoft's server software has now claimed about 400 victims, according to researchers at Netherlands-based Eye Security. The figure, which is derived from a count of digital artifacts discovered during scans of servers running vulnerable versions of Microsoft's SharePoint software, compares to 100 organizations cataloged over the weekend. Eye Security says the figure is likely an undercount. "There are many more, because not all attack vectors have left artifacts that we could scan for," said Vaisha Bernard, the chief hacker for Eye Security, which was among the first organizations to flag the breaches. The spy campaign kicked off after Microsoft failed to fully patch a security hole in its SharePoint server software, kicking off a scramble to fix the vulnerability when it was discovered. Microsoft and its tech rival, Google owner Alphabet GOOGL.O, have both said Chinese hackers are among those taking advantage of the flaw. Beijing has denied the claim. The details of most of the victim organizations have not yet been fully disclosed. Bernard declined to identify them. — Reuters
GMA Network
a day ago
- GMA Network
Humans beat AI gold-level score at top math contest
SYDNEY — Humans beat generative AI models made by Google and OpenAI at a top international mathematics competition, despite the programs reaching gold-level scores for the first time. Neither model scored full marks—unlike five young people at the International Mathematical Olympiad (IMO), a prestigious annual competition where participants must be under 20 years old. Google said Monday that an advanced version of its Gemini chatbot had solved five out of the six math problems set at the IMO, held in Australia's Queensland this month. "We can confirm that Google DeepMind has reached the much-desired milestone, earning 35 out of a possible 42 points—a gold medal score," the US tech giant cited IMO president Gregor Dolinar as saying. "Their solutions were astonishing in many respects. IMO graders found them to be clear, precise and most of them easy to follow." Around 10 percent of human contestants won gold-level medals, and five received perfect scores of 42 points. US ChatGPT maker OpenAI said that its experimental reasoning model had scored a gold-level 35 points on the test. The result "achieved a longstanding grand challenge in AI" at "the world's most prestigious math competition," OpenAI researcher Alexander Wei wrote on social media. "We evaluated our models on the 2025 IMO problems under the same rules as human contestants," he said. "For each problem, three former IMO medalists independently graded the model's submitted proof." Google achieved a silver-medal score at last year's IMO in the British city of Bath, solving four of the six problems. That took two to three days of computation—far longer than this year, when its Gemini model solved the problems within the 4.5-hour time limit, it said. The IMO said tech companies had "privately tested closed-source AI models on this year's problems," the same ones faced by 641 competing students from 112 countries. "It is very exciting to see progress in the mathematical capabilities of AI models," said IMO president Dolinar. Contest organizers could not verify how much computing power had been used by the AI models or whether there had been human involvement, he cautioned. — Agence France-Presse
GMA Network
5 days ago
- GMA Network
Singapore says cyber espionage group targeting critical infrastructure
A view of the central business district skyline in Singapore May 27, 2025. REUTERS/ Edgar Su SINGAPORE - Singapore said on Friday that it was responding to cyberattacks on its critical infrastructure by an espionage group alleged by security experts to be linked to China. "UNC3886 poses a serious threat to us, and has the potential to undermine our national security,' Coordinating Minister for National Security K. Shanmugam said in a speech. "It is going after high value strategic threat targets, vital infrastructure that delivers essential services." He did not give details of the attacks, citing security risks, nor of any consequences. Google-owned cybersecurity firm Mandiant has described UNC3886 as a "China-nexus espionage group" that has attacked defense, technology and telecommunications organizations in the US and Asia. Beijing routinely denies any allegations of cyberespionage, and says it opposes all forms of cyberattacks and is in fact a victim of such threats. The Chinese embassy did not immediately respond to a request for comment sent after office hours. Singapore's critical infrastructure sectors include energy, water, banking, finance, healthcare, transport, government, communication, media, as well as security and emergency services, according to the country's cyber agency. Reuters earlier this week reported that the Taiwanese semiconductor industry and investment analysts had been targeted by Chinese-linked hackers as part of a string of cyber espionage campaigns. —Reuters
