# Experts creating 'Humanity's Last Exam' to stump powerful AI tech | Tech News (HT Tech) > Dubbed Humanity's Last Exam, the project seeks to determine when expert-level AI has arrived. # Experts creating 'Humanity's Last Exam' to stump powerful AI tech ## Dubbed "Humanity's Last Exam," the project seeks to determine when expert-level AI has arrived. Sept 17 2024, 12:50 IST ![Follow on Facebook](https://images.hindustantimes.com/tech/static/revamp/images/png/follow-us-facebook.png) ![Follow on Facebook](https://images.hindustantimes.com/tech/static/revamp/images/png/follow-us-facebook-hover.png) ![Share](https://images.hindustantimes.com/tech/static/revamp/images/png/Share.png) ![Share](https://images.hindustantimes.com/tech/static/revamp/images/png/Shareclicked.png) ![Share on WhatsApp](https://images.hindustantimes.com/tech/static/revamp/images/png/whatsapp.png) ![Share on WhatsApp](https://images.hindustantimes.com/tech/static/revamp/images/png/whatsapp-hover.png) ![Share on Twitter](https://images.hindustantimes.com/tech/static/revamp/images/png/twitter.png) ![Share on Twitter](https://images.hindustantimes.com/tech/static/revamp/images/png/twitter-hover.png) ![Share on Facebook](https://images.hindustantimes.com/tech/static/revamp/images/png/facebook.png) ![Share on Facebook](https://images.hindustantimes.com/tech/static/revamp/images/png/facebook-hover.png) ![Bookmark](https://images.hindustantimes.com/tech/static/revamp/images/png/Bookmark.png) ![Bookmarked](https://images.hindustantimes.com/tech/static/revamp/images/png/Bookmarked.png) ![Experts creating 'Humanity's Last Exam' to stump powerful AI tech](https://images.hindustantimes.com/tech/img/2024/09/17/1600x900/HEALTH-AI-WHITE-HOUSE-0_1703814543854_1726557433884.JPG) AI has appeared to score poorly on lesser-used tests involving plan formulation and visual pattern-recognition puzzles, according to Stanford University’s AI Index Report from April. (REUTERS) ![Author Photo](https://images.hindustantimes.com/tech/static/revamp/images/png/HTTech-new-circle-logo.png) ### By REUTERS A team of technology experts issued a global call on Monday seeking the toughest questions to pose to artificial intelligence systems, which increasingly have handled popular benchmark tests like child's play. Dubbed "Humanity's Last Exam," the project seeks to determine when expert-level AI has arrived. It aims to stay relevant even as capabilities advance in future years, according to the organisers, a non-profit called the Center for AI Safety (CAIS) and the startup Scale AI. The call comes days after the maker of ChatGPT previewed a new model, known as OpenAI o1, which "destroyed the most popular reasoning benchmarks," said Dan Hendrycks, executive director of CAIS and an advisor to Elon Musk's xAI startup. Also read click here. Hendrycks co-authored two 2021 papers that proposed tests of AI systems that are now widely used, one quizzing them on undergraduate-level knowledge of topics like U.S. history, the other probing models' ability to reason through competition-level math. The undergraduate-style test has more downloads from the online AI hub Hugging Face than any such dataset. At the time of those papers, AI was giving almost random answers to questions on the exams. "They're now crushed," Hendrycks told Reuters. As one example, the Claude models from the AI lab Anthropic have gone from scoring about 77% on the undergraduate-level test in 2023, to nearly 89% a year later, according to a prominent capabilities leaderboard. These common benchmarks have less meaning as a result. AI has appeared to score poorly on lesser-used tests involving plan formulation and visual pattern-recognition puzzles, according to Stanford University’s AI Index Report from April. OpenAI o1 scored around 21% on one version of the pattern-recognition ARC-AGI test, for instance, the ARC organizers said on Friday. Some AI researchers argue that results like this show planning and abstract reasoning to be better measures of intelligence, though Hendrycks said the visual aspect of ARC makes it less suited to assessing language models. “Humanity’s Last Exam” will require abstract reasoning, he said. Answers from common benchmarks may also have ended up in data used to train AI systems, industry observers have said. Hendrycks said some questions on "Humanity's Last Exam" will remain private to make sure AI systems' answers are not from memorisation. The exam will include at least 1,000 crowd-sourced questions due November 1 that are hard for non-experts to answer. These will undergo peer review, with winning submissions offered co-authorship and up to $5,000 prizes sponsored by Scale AI. “We desperately need harder tests for expert-level models to measure the rapid progress of AI," said Alexandr Wang, Scale's CEO. One restriction: the organisers want no questions about weapons, which some say would be too dangerous for AI to study.One more thing! We are now on WhatsApp Channels! Follow us there so you never miss any updates from the world of technology. ‎To follow the HT Tech channel on WhatsApp, click here to join now! Catch all the Latest Tech News, Mobile News, Laptop News, Gaming news, Wearables News , How To News, also keep up with us on Whatsapp channel,Twitter, Facebook, Google News, and Instagram. For our latest videos, subscribe to our YouTube channel. 17 Sept, 12:50 IST artificial intelligence ![HT Tech Awards](https://tech.hindustantimes.com/ht-tech-power-list-awards-2026/httech-awards-banner-new.webp) ### Related Stories 07 Oct 2026 71,726,556,597,900 NEXT ARTICLE BEGINS ![View all](https://images.hindustantimes.com/tech/static/revamp/images/png/view-all-icon.png) ## Tips & Tricks ![Ghibli-style art](https://images.hindustantimes.com/tech/img/2025/03/28/414x233/ChatGPT_Ghibli_style_images_1743161814450_1743161814596.jpg) ### How to turn your photos into Ghibli-style art with ChatGPT for free: Step-by-step guide ![How to create an invitation using iPhone’s new Apple Invites app- Step-by-step guide](https://images.hindustantimes.com/tech/img/2025/02/14/414x233/invites_1738728829564_1739543695649.png) ### How to create an invitation using iPhone’s new Apple Invites app- Step-by-step guide ![Apple TV app arrives on Android phones and tablets: What you get and how to install it](https://images.hindustantimes.com/tech/img/2025/02/13/414x233/unnamed_1739426557335_1739426567634.jpg) ### Apple TV app arrives on Android phones and tablets: What you get and how to install it ![Vodafone Idea OTT plans](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### Vodafone Idea users can now stream 17 OTT apps with one subscription: Here's how ![Android data transfer methods](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### How to transfer data from one Android to another using Google backup, USB-C cable, and more ## Editor’s Pick ![iPhone 16e launched with A18 chip, Apple AI—bringing flagship power on a budget: Check specs, price](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### iPhone 16e launched in India: All details about the new budget mobile from Apple ![DeepSeek was founded in China with only a fraction of the budget of the top AI companies in the US.](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### DeepSeek AI: How it works, who’s behind it, how it’s different from ChatGPT, and its big US market impact explained ![iPhone 17 Air likely to get this feature that iPhone 17 Pro models may miss out on](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### iPhone 17 Air likely to get this feature that iPhone 17 Pro models may miss out on ![Google Whisk AI explained: How remixing works, availability, and how it differs from Gemini](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### Google Whisk AI explained: How remixing works, availability, and how it differs from Gemini ![iOS 18.4 beta arriving soon: Know why it is the biggest iPhone update](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### iOS 18.4 beta arriving soon: Know why it is the biggest iPhone update ## Trending Stories ![Airtel 5G SIM card will now be delivered in 10 minutes with Blinkit- How to order](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### Airtel 5G SIM card will now be delivered in 10 minutes with Blinkit- How to order ![Infinix Note 50s 5G+](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### Infinix Note 50s 5G+ set to launch in India with 64MP camera and MediaTek Dimensity 7300 SoC ![Xiaomi X Pro QLED Series sale starts in India: Check price, availability, and more](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### Xiaomi X Pro QLED Series sale starts in India: Check price, availability, and more ![Apple Intelligence AI training](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### Apple plans to train AI using your on-device data without accessing personal content or emails: Report ![CMF Buds 2](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### CMF Buds 2 key features and price revealed ahead of April 28 launch: Details inside keep up with tech ![Facebook](https://images.hindustantimes.com/tech/static/revamp/images/png/tech-facebook.png) ![Twitter](https://images.hindustantimes.com/tech/static/revamp/images/png/tech-twitter.png) ![WhatsApp](https://images.hindustantimes.com/tech/static/revamp/images/png/tech-whatsApp.png) ![Facebook](https://images.hindustantimes.com/tech/static/revamp/images/png/tech-facebook.png) ## Gaming ![Check out these five gaming monitors under ₹20,000,](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### Top 5 gaming monitors under ₹20000 you should consider ![gaming consoles](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### Grab gaming consoles at discounted rates during Amazon Great Indian Festival: Up to 72% off ![Pokémon GO Wild Area 2025](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### Pokémon GO Wild Area 2025 to bring Grimmsnarl family, Shiny Hatenna, and global challenges ![Shade Silver](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### Shade Silver free on Steam for a limited time: Here’s how to get it ![Call of duty](https://images.hindustantimes.com/tech/static/revamp/images/png/ht-tech-bg-img-90-66.png) ### Call of Duty: Black Ops 7 skips Switch 2 at launch, release date and price leak ![View all](https://images.hindustantimes.com/tech/static/revamp/images/png/view-all-icon.png) ## Trending News ### Airtel 5G SIM card will now be delivered in 10 minutes with Blinkit- How to order ![Airtel 5G SIM card will now be delivered in 10 minutes with Blinkit- How to order](https://images.livemint.com/static/static/1x1_img.gif) ### Infinix Note 50s 5G+ set to launch in India with 64MP camera and MediaTek Dimensity 7300 SoC ![Infinix Note 50s 5G+](https://images.livemint.com/static/static/1x1_img.gif) ### Xiaomi X Pro QLED Series sale starts in India: Check price, availability, and more ![Xiaomi X Pro QLED Series sale starts in India: Check price, availability, and more](https://images.livemint.com/static/static/1x1_img.gif) ### Apple plans to train AI using your on-device data without accessing personal content or emails: Report ![Apple Intelligence AI training](https://images.livemint.com/static/static/1x1_img.gif) ### CMF Buds 2 key features and price revealed ahead of April 28 launch: Details inside ![CMF Buds 2](https://images.livemint.com/static/static/1x1_img.gif) ## Trending Gadgets Mobiles Laptops Tablets ![HT Tech](https://images.hindustantimes.com/tech/static/revamp/images/png/HT_tech-new.gif) - https://tech.hindustantimes.com/tech/news/experts-creating-humanitys-last-exam-to-stump-powerful-ai-tech-71726556597900.html