Latest news with #AlphaZero
&w=3840&q=100)
First Post
5 days ago
- Science
- First Post
Which AI model is best? Big chess battle between top engines to decide
Eight top AI models from Google, OpenAI, Anthropic, and more are battling it out in the Kaggle AI Chess Exhibition Tournament to find out which model is best at chess. Day 1 saw dominant 4-0 wins across all games. The semi-finals will be held on Wednesday, August 6 and the final is slated for Thursday. read more World's top AI models are competing in a chess exhibition tournament to find out who is the best. AFP A new and exciting tournament is being held to find out which AI model is the best at playing chess. The event is being organised by Kaggle Game Arena in partnership with Google DeepMind. It is called the AI Chess Exhibition Tournament, and is being held from August 5 to August 7. This special tournament will feature eight popular Large Language Models (LLMs) from top companies like Google, OpenAI, Anthropic, and more. These models are not made for chess, but for other things like coding, writing, and answering questions. However, they now face each other on the chessboard. STORY CONTINUES BELOW THIS AD The competing AIs and format Gemini 2.5 Pro (Google) Gemini 2.5 Flash (Google) o3 (OpenAI) o4-mini (OpenAI) Claude 4 Opus (Anthropic) Grok 4 (xAI, Elon Musk's company) DeepSeek R1 Kimi k2 (Moonshot AI) The tournament will follow a knockout format, which means one loss and the model is out. Based on how they play, the AIs will be ranked, and the results will be shared on and on Kaggle. American Grandmaster Hikaru Nakamura is also covering the event on his Twitch handle. AI Chess Exhibition Tournament Day 1 results Day 1 of the Kaggle AI Chess Tournament saw dominant performances, with four models, OpenAI o4 mini, OpenAI o3, Gemini 2.5 Pro, and Grok 4, cruising into the semifinals with identical 4-0 wins. OpenAI's models impressed early, as both o4 mini and o3 swept past DeepSeek-R1 and Kimi K2 Instruct respectively. On the other side of the bracket, Google's Gemini 2.5 Pro outplayed Anthropic's Claude Opus 4, while xAI's Grok 4 defeated Gemini 2.5 Flash in a surprise result. In the semifinals on August 6, o4 mini will face o3, and Gemini 2.5 Pro will lock horns against Grok 4. The gold-medal match and bronze-medal match will be held on August 7. In 2017, Google's famous AI model AlphaZero shocked the world when it learned chess in just 4 hours and beat Stockfish, which is considered the world's best chess engine. But unlike AlphaZero, the current models are not made for chess as they are general-purpose language models. However, it will still be an interesting tournament to watch, as it will provide an answer to which AI model is the best at chess.
Indian Express
6 days ago
- Entertainment
- Indian Express
Which is the smartest AI model? A chess tournament might hold the answer
Starting from Tuesday, a unique chess tournament might offer the answer to the question on everyone's mind: which is the smartest artificial intelligence (AI) model. The chess tournament, to be played over three days on Kaggle Game Arena, in partnership with Google DeepMind. The tournament will also have world no 2 chess player Hikaru Nakamura doing livestreams of the action on his Twitch channel with his insights while world no 1 Magnus Carlsen will do an event-ending recap for the Take Take Take app. Popular chess streamer Levy Rozman (known popularly as Gotham Chess) will also be doing daily recaps and analysis videos on his YouTube channel. The event will feature eight of the world's leading AI models: Gemini 2.5 Pro (Google), Gemini 2.5 Flash (Google), o3 (OpenAI), o4-mini (OpenAI), Claude 4 Opus (Anthropic), Grok 4 (xAI), DeepSeek R1 and Kimi k2 (Moonshot AI). Elon Musk, who is the man behind Grok, was slammed by chess players last year for his opinion about chess. 'Computers are so much better than humans at chess, it's absurd. I predict that chess will be essentially fully solved (like checkers) within 10 years,' he had posted on X in May last year. He had also spoken in an interview about playing chess for his school chess team. 'I was on the school chess team. But I find that chess is a simple game frankly. I mean you only have 64 squares. I was pretty good at chess as a kid. I won every game.' Responding to Musk's comments about never losing at chess and it being a simple sport, Nakamura had said: 'Anybody who says they had never lost at chess, you know something (fishy) is going on! Back in the 90s machine vs human and machine vs machine chess tournaments (such as the Top Chess Engine Championship or Computer Chess Championship) used to be quote popular. But after the machines grew too strong, the events ran out of steam and were discontinued. The event will be help with help from popular streaming platform and Google DeepMind, which gave the world the chess-playing computer program AlphaZero in 2017. AlphaZero is a neural network that became really strong in the sport by playing millions of games against itself for four hours. In fact, AlphaZero was able to outperform Stockfish 8, which was back then one of the strongest chess programs, just four hours after being fed the rules of chess and being told to learn by playing simulations against itself. When AlphaZero and Stockfish played a 100-game match, AlphaZero won or drew all of the games.
Time of India
29-07-2025
- Business
- Time of India
Tesla CEO Elon Musk says he stopped playing chess as a kid, but agrees it has become ‘more popular and bigger'
Tesla and SpaceX CEO Elon Musk recently said that he mostly stopped playing chess as a child after realizing that computers would one day easily defeat human players. The tech billionaire was responding to a post by American businessman Marc Andreessen who wrote: 'Chess is more popular and a bigger industry now than ever.' Andreessen's post highlights chess's resurgence, contradicting fears of AI-driven obsolescence, with global participation hitting 605 million players in 2022 per FIDE's data, a 30% rise since 2012, fueled by online platforms like during the COVID-19 pandemic. To this, Elon Musk replied 'True, to my surprise. I mostly stopped playing chess as a kid when I realized that it would be trivial in the future to write a chess (low DoF game) program that could beat all humans.' In this context, 'low DoF' refers to 'low degrees of freedom,' meaning the game has a limited set of possible moves and outcomes, making it easier for a computer to master through programming. In 1997, IBM's Deep Blue made history by defeating world champion Garry Kasparov in a six-game match, becoming the first computer to beat a reigning world chess champion under standard tournament conditions. Since then, chess engines have only become more powerful. Modern AI systems like Stockfish and Google's AlphaZero can calculate millions of positions per second, easily outperforming even the strongest human players. AlphaZero, developed by DeepMind (a Google-owned company), learned the game from scratch using reinforcement learning and went on to beat top engines using unconventional and creative strategies — a milestone in AI development. Despite AI dominance in the game, chess has seen a global resurgence thanks to online platforms, live-streamed tournaments, and popular culture moments like Netflix's The Queen's Gambit. Musk acknowledged this rise in popularity but maintained that the game's eventual solvability by machines made it less appealing to him. iQOO Z10R 5G goes on Sale: BEST Budget Phone for Content Creators? AI Masterclass for Students. Upskill Young Ones Today!– Join Now
Time Magazine
29-07-2025
- Time Magazine
Is 'Sweatshop Data' Really Over?
Welcome back to In the Loop, TIME's new twice-weekly newsletter about the world of AI. If you're reading this in your browser, you can subscribe to have the next one delivered straight to your inbox. What to Know: The future of 'sweatshop data' You can measure time in the world of AI by the cadence of new essays with provocative titles. Another one arrived earlier this month from the team at Mechanize Work: a new startup that is trying to, er, automate all human labor. Its title? 'Sweatshop data is over.' This one caught my attention. As regular readers may know, I've done a lot of reporting over the years on the origins of the data that is used to train AI systems. My story 'Inside Facebook's African Sweatshop' was the first to reveal how Meta used contractors in Kenya, some earning as little as $1.50 per hour, to remove content from their platforms—content that would later be used in attempts to train AI systems to do that job automatically. I also broke the news that OpenAI used workers from the same outsourcing company to detoxify ChatGPT. In both cases, workers said the labor left them with diagnoses of post-traumatic stress disorder. So if sweatshop data really is a thing of the past, that would be a very big deal indeed. What the essay argues — Mechanize Work's essay points to a very real trend in AI research. To summarize: AI systems used to be relatively unintelligent. To teach them the difference between, say, a cat and a dog, you'd need to give them lots of different labeled examples of cats and dogs. The most cost-effective way to get those labels was from the Global South, where labor is cheap. But as AI systems have gotten smarter, they no longer need to be told basic information, the authors argue. AI companies are now desperately seeking expert data, which necessarily comes from people with PhDs—and who won't put up with poverty wages. 'Teaching AIs these new capabilities will require the dedicated efforts of high-skill specialists working full-time, not low-skill contractors working at scale,' the authors argue. A new AI paradigm — The authors are, in one important sense, correct. The big money has indeed moved toward expert data. A clutch of companies, including Mechanize Work, are jostling to be the ones to dominate the space, which could eventually be worth hundreds of billions of dollars, according to insiders. Many of them aren't just hiring experts, but are also building dedicated software environments to help AI learn from experience at scale, in a paradigm known as reinforcement learning with verifiable rewards. It takes inspiration from DeepMind's 2017 model AlphaZero, which didn't need to observe humans playing chess or Go, and instead became superhuman just by playing against itself millions of times. In the same vein, these companies are trying to build software that would allow AI to 'self-play,' with the help of experts, on questions of coding, science, and math. If they can get that to work, it could potentially unlock major new leaps in capability, top researchers believe. There's just one problem — While all of this is true, it does not mean that sweatshop data has gone away. 'We don't observe the workforce of data workers, in the classical sense, decreasing,' says Milagros Miceli, a researcher at the Weizenbaum Institute in Berlin who studies so-called sweatshop data. 'Quite the opposite.' Meta and TikTok, for example, still rely on thousands of contractors all over the world to remove harmful content from their systems—a task that has stubbornly resisted full AI automation. Other types of low-paid tasks, typically carried out in places like Kenya, the Philippines, and India, are booming. 'Right now what we are seeing is a lot of what we call algorithmic verification: people checking in on existing AI models to ensure that they are functioning according to plan,' Miceli says. 'The funny thing is, it's the same workers. If you talk to people, they will tell you: I have done content moderation. I have done data labeling. Now I am doing this.' Who to Know: Shengjia Zhao, Chief Scientist, Meta Superintelligence Labs Mark Zuckerberg promoted AI researcher Shengjia Zhao to chief scientist of the new effort inside Meta to create 'superintelligence.' Zhao joined Meta last month from OpenAI, where he worked on the o1-mini and o3-mini models. Zuck's memo — In a note to staff on Saturday, Zuckerberg wrote: 'Shengjia has already pioneered several breakthroughs including a new scaling paradigm and distinguished himself as a leader in the field.' Zhao, who studied for his undergraduate degree in Beijing and graduated from Stanford with a PhD in 2022, 'will set the research agenda and scientific direction for our new lab,' Zuckerberg wrote. Meta's recruiting push — Zuckerberg has ignited a fierce war for talent in the AI industry by offering top AI researchers pay packages worth up to $300 million, according to reports. 'I've lost track of how many people from here they've tried to get,' Sam Altman told OpenAI staff in a Slack message, according to the Wall Street Journal. Bad news for LeCun — Zhao's promotion is yet another sign that Yann LeCun—who until the hiring blitz this year was Meta's most senior AI scientist—has been put out to pasture. A notable critic of the idea that LLMs will scale to superintelligence, LeCun's views appear to be increasingly at odds with Zuckerberg's bullishness. Meta's Superintelligence team is clearly now a higher priority for Zuckerberg than the separate group LeCun runs, called Facebook AI Research (FAIR). In a note appended to his announcement of Zhao's promotion on Threads, Zuckerberg denied that LeCun had been sidelined. 'To avoid any confusion, there's no change in Yann's role,' he wrote. 'He will continue to be Chief Scientist for FAIR.' AI in Action One of the big ways AI is already affecting our world is in the changes it's bringing to our information ecosystem. News publishers have long complained that Google's 'AI Overviews' in its search results have reduced traffic, and therefore revenues, harming their ability to employ journalists and hold the powerful to account. Now we have new data from the Pew Research Center that puts that complaint into stark relief. When AI summaries are included in search results, only 8% of users click through to a link — down from 15% without an AI summary, the study found. Just 1% of users clicked on any link in that AI summary itself, rubbishing the argument that AI summaries are an effective way of sending users toward publishers' content. As always, if you have an interesting story of AI in Action, we'd love to hear it. Email us at: intheloop@ What We're Reading 'How to Save OpenAI's Nonprofit Soul, According to a Former OpenAI Employee,' by Jacob Hilton in TIME Jacob Hilton, who worked at OpenAI between 2018 and 2023, writes about the ongoing battle over OpenAI's legal structure—and what it might mean for the future of our world. 'The nonprofit still has no independent staff of its own, and its board members are too busy running their own companies or academic labs to provide meaningful oversight,' he argues. 'To add to this, OpenAI's proposed restructuring now threatens to weaken the board's authority when it instead needs reinforcing.'
Yahoo
10-02-2025
- Business
- Yahoo
'Important' to regulate AI but it's 'tricky': Google DeepMind CEO
STORY: :: Google Deepmind CEO says regulating AI is 'important' but 'tricky in the current environment' :: February 9, 2025 :: Paris, France :: Demis Hassabis, Nobel Laureate and Google Deepmind CEO 'I think it's important to regulate AI., but it's important to get the regulations right and that's hard, when such that the technology itself is not fully understood and it's so fast moving. And it also needs to be international because AI is going to affect all countries, the whole world, as a technology, it needs to be... there needs to be sort of international cooperation around that. And that's also tricky in the current environment." 'I mean, just briefly on DeepSeek, it's an impressive piece of work and I think it's probably the best work I've seen come out of China. But it's important to understand that, despite the hype, there's no actual new scientific advance there. It's using known techniques. Actually, many of the techniques we invented at Google, and at DeepMind, things like AlphaZero and some of the reinforcement learning, they use." The AI Action Summit, which kicks off on Monday (February 10) at Paris' Grand Palais, will discuss how to safely embrace artificial intelligence at a time of mounting resistance to heavy-handed red tape that businesses say stifles innovation. The summit will also aim to find common ground between U.S. President Donald Trump's administration, China and nearly 100 other nations on how to regulate AI. An official for the French presidency said the summit will give voice to countries around the world, not only the likes of U.S. and China. Mitigating labor disruption and promoting sovereignty in a global AI market are also on the agenda. Google Senior Vice President James Manyika said global conversations on AI were usually "focused on the risks and complexities and not enough on the opportunities," citing AI's potential to help developing countries in education, climate change and health care. A non-binding communiqué of principles for the stewardship of AI, bearing U.S., Chinese and other signatures, has been under negotiation and would mark a big achievement if reached, said the people involved in the summit, who spoke on condition of anonymity. Instead in focus is how to distribute AI's benefits to developing nations, via cheaper models made by the likes of France's startup Mistral and China's DeepSeek. The Hangzhou-based company rocked global markets last month by showing it could vie with U.S. heavyweights on human-like reasoning technology, at lower costs. Hannabis said the hype over DeepSeek was "a little bit exaggerated" as there was "no actual new scientific advance", but said it demonstrated "extremely good engineering". Sign in to access your portfolio
