logo
Google and OpenAI's AI models win milestone gold at global math competition

Google and OpenAI's AI models win milestone gold at global math competition

The Hindu22-07-2025
Alphabet's Google and OpenAI said their artificial-intelligence models won gold medals at a global mathematics competition, signaling a breakthrough in math capabilities in the race to build powerful 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 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.
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 specialised 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. IMO judges certified the results of those companies, including Google, and asked them to publish results on July 28.
"We respected the IMO Board's original request that all AI labs share their results only after the official results had been verified by independent experts and the students had rightly received the acclamation they deserved," Google DeepMind CEO Demis Hassabis said on X on Monday.
OpenAI, which published its results on Saturday and first claimed gold-medal status, said in an interview that it had permission from an IMO board member to do so after the closing ceremony on Saturday.
The competition on Monday allowed cooperating companies to publish results, Gregor Dolinar, president of IMO's board, told Reuters.
Orange background

Try Our AI Features

Explore what Daily8 AI can do for you:

Comments

No comments yet...

Related Articles

Indian Woman Who Was Told She'll Never Make It To Google Now Works At The Tech Giant. See Viral Post
Indian Woman Who Was Told She'll Never Make It To Google Now Works At The Tech Giant. See Viral Post

NDTV

time27 minutes ago

  • NDTV

Indian Woman Who Was Told She'll Never Make It To Google Now Works At The Tech Giant. See Viral Post

An Indian woman is making headlines after sharing about a past interview experience where she was mocked by a startup interviewer who claimed she would never make it to global tech giants like Google or Meta. Taking to X, the user who goes by Arpita Das on the microblogging platform, narrated her interview experience, revealing that she eventually did get into Google. "Was grilled by a mid-level startup interviewer in a system design round, he made me design infra, estimate CPU costs, basically everything except physically build the data centre," she wrote in her post. However, she said that the interview took a sharp turn when she struggled to answer a few questions. "When I stumbled, he smirked and said, 'This is why people like you won't make it to big companies like Google, Meta'". But proving the interviewer wrong, Ms Das revealed that she now works at Google. "Not bragging-just wondering why some folks gatekeep based on their own insecurities," she concluded. Take a look at the post below: Was grilled by a mid-level startup interviewer in a system design round - he made me design infra, estimate CPU costs, basically everything except physically build the data center. When I stumbled, he smirked and said, 'This is why people like you won't make it to big companies… — Arpita Das (@Arpitaaa01) July 27, 2025 The post quickly caught the internet's attention. While many praised her success story, others slammed the interviewer. Some users also shared similar experiences. "Great story. Success is the best revenge. Keep going," one user wrote. "I think in later stages you will get the chance to interview that interviewer in future, that's how the world works," joked another. "Today, most interviewers see attitude and eagerness to learn as you can not judge a person on the whole thing in those 10-15 minutes," said a third user. "The person who was reluctant to hire me during the interview was fired after 1.5 years. He thought that, as a woman, I won't be able to do onsite work. However, I ended up performing not only my own duties but also his, and I did so better than he did. He had become redundant," shared one user. "Hiring system is terribly messed up nowadays and I'm pretty sure the startup is in wrong you got exactly what you deserve," commented another. Notably, according to her X bio, Ms Das has previously worked at prominent companies such as Microsoft and Myntra.

OpenAI CEO Sam Altman flags lack of AI-user privacy; Deepfakes threaten political security; Nvidia orders H20 chips from TSMC
OpenAI CEO Sam Altman flags lack of AI-user privacy; Deepfakes threaten political security; Nvidia orders H20 chips from TSMC

The Hindu

time27 minutes ago

  • The Hindu

OpenAI CEO Sam Altman flags lack of AI-user privacy; Deepfakes threaten political security; Nvidia orders H20 chips from TSMC

OpenAI CEO Sam Altman flags lack of AI-user privacy Users who treat ChatGPT as a life coach or therapist, sharing extremely personal thoughts or sensitive questions with the AI chatbot, do not have real privacy protections, according to OpenAI CEO Sam Altman. The technologist warned during an episode of 'This Past Weekend' podcast with Theo Von that the AI company could be forced to hand over such content as evidence in case of a lawsuit or a legal proceeding. Altman made it clear that he disagreed with this standard and highlighted the need for a user-AI privacy standard similar to the one that exists between a patient and a doctor or a client and a therapist. Such interactions are generally protected by what is known as 'privilege,' meaning it is hard to use them against the individual in court. This is not merely a hypothetical situation for the company and its users worldwide. OpenAI was critical of The New York Times, claiming that as part of a lawsuit against the AI startup, the media company asked the court to force OpenAI to 'retain all user content indefinitely going forward...' Deepfakes threaten political security Employees who work in U.S. President Donald Trump's administration, high-profile business companies in America, or corporate settings are preparing to tackle the surge in realistic deepfakes across media formats that impersonate their leaders and chiefs. This has been done to U.S. Secretary of State Marco Rubio as well as Trump's chief of staff, Susie Wiles. There is a serious risk that lower-level employees targeted by such scams might disclose highly sensitive information, security details, or trade secrets to malicious attackers representing hostile regimes. Deepfakes can also be used to scam users out of their savings, get access to their passwords, or prevent them from exercising their right to vote unimpeded. However, there are tech experts who believe that the solution to identifying AI-generated deepfakes is deploying more AI tools that can process the details of a deepfake that human viewers might miss out on. Efforts are also on to better track North Korean agents who are embedded in the U.S. workforce. Nvidia orders H20 chips from TSMC Sources have claimed that Nvidia ordered 300,000 H20 chipsets from TSMC last week, with the development being related to a strong Chinese demand for the chips that are used in AI development. H20 was developed by Nvidia for the Chinese market and has less computing power than Nvidia's H100 and its new Blackwell series; it was made after U.S. export restrictions hit its other AI chipsets. Nvidia and TSMC have not yet formally commented on the reports. While U.S. legislators are concerned that China's access to the H20 offerings will affect competition between the two countries when it comes to leading AI technology, Nvidia's perspective is that Chinese interest in its chips will prevent developers from becoming reliant on products from rivals such as Huawei.

Beyond Gemini, Samsung Galaxy S26 5G will likely let you pick its AI brain
Beyond Gemini, Samsung Galaxy S26 5G will likely let you pick its AI brain

Hindustan Times

time40 minutes ago

  • Hindustan Times

Beyond Gemini, Samsung Galaxy S26 5G will likely let you pick its AI brain

As Samsung prepares for next year's flagship launch, it is reported to expand its AI reach for its upcoming devices. While we are accumulating the Galaxy S26 5G series leaks and rumours, Samsung is reportedly in talks with several companies to expand its AI services apart from Google's Gemini. Therefore, it is expected that the upcoming Samsung Galaxy S26 series could offer additional AI assistant options, instead of Bixby and Gemini. Samsung is expected to finalise a deal with Perplexity for integrating its AI chatbot. Here's what we know so far.(Bloomberg) Now, Samsung users will have greater chatbot support, as they will be able to decide with an AI assistant they want to use on a day-to-day basis. Samsung Mobile Division President and COO, Choi Won-Joon, has also confirmed the claims of the company 'talking to multiple vendors' for AI integration. Therefore, know what AI upgrades we can expect from the Galaxy S26 series Samsung Galaxy S26 5G series likely to get a major AI upgrade According to a Bloomberg report, Samsung is in talks with multiple AI companies to expand its AI capabilities for the upcoming Galaxy S26 5G series. It was highlighted that the upcoming flagship could see options for several AI assistants apart from Google's Gemini. Regarding the expansion, Choi Won-Joon said, 'As long as these AI agents are competitive and can provide the best user experiences, we are open to any AI agent out there.' Earlier, Samsung was also rumoured to be in talks with Perplexity for its AI chatbot expansion. Now, the recent report has stated that the companies are nearing finalisation. Samsung is also expected to invest in Perplexity while also bringing its AI capabilities to its smartphones. Now, users will not have to rely on any specific AI assistant for their AI interactions. If the deal pans out, these new AI assistants will be preinstalled on the Galaxy S26 series, similar to how we access Gemini. However, Samsung is yet to finalise the deal, and may reach out to other vendors as well before the launch. With AI expansion, Samsung is not only expanding its AI reach but also reducing its dependence on Google. Additionally, if the deal went through, Samsung's flagship models would also compete with Apple iPhones, since they rely on OpenAI's ChatGPT for AI tasks and search queries. Apart from this AI upgrade, the Samsung Galaxy S26 Ultra is expected to get a major camera and performance boost. Reportedly, the Galaxy S26 series is expected to be powered by a Samsung-made Snapdragon 8 Elite 2 chip. We may also get an upgrade to 16GB of RAM next year.

DOWNLOAD THE APP

Get Started Now: Download the App

Ready to dive into a world of global content with local flavor? Download Daily8 app today from your preferred app store and start exploring.
app-storeplay-store