Loading...
Market
What will be the IQ score of the next version of OpenAI's o1 model by June 30, 2025?
2
Resolution / Starting Odds
Below 120 • 25%
120-130 • 25%
130-140 • 25%
Above 140 • 25%
Official announcements from OpenAI or validated IQ tests
Story
OpenAI Unveils New o1 Model with Advanced Reasoning, IQ Score of 120
Sep 16, 2024, 12:20 AM
OpenAI has released a new AI model named 'o1', also known as 'Strawberry'. This model, which is distinct from previous iterations like ChatGPT, is designed to perform advanced reasoning tasks. It has demonstrated significant improvements over its predecessors, such as GPT-4o, particularly in complex applications including software programming, STEM, legal, disease diagnosis, and scientific research. The o1 model has achieved an IQ score of 120, outperforming most humans and other AI models. It uses a chain-of-thought reasoning approach and can fact-check itself. The model scored 25 out of 35 on the Mensa IQ test, demonstrating PhD-level intelligence. Despite its impressive capabilities, there are concerns about its potential misuse, including the creation of bioweapons. OpenAI plans to continue enhancing the model, with a new version expected in early 2025.
View original story
Similar markets
Yes • 50%
No • 50%
None • 25%
1 to 2 • 25%
3 to 4 • 25%
5 or more • 25%
Reducing hallucinations • 25%
Improving multilingual capabilities • 25%
Enhancing CoT reasoning • 25%
Other • 25%
Reinforcement learning • 25%
Search-based reasoning • 25%
Thinking before answering • 25%
Other • 25%
Improved Natural Language Understanding • 25%
Enhanced Coding Abilities • 25%
Advanced Scientific Research Capabilities • 25%
Other • 25%
Yes • 50%
No • 50%
Yes • 50%
No • 50%
Less than 1 million • 25%
1-5 million • 25%
5-10 million • 25%
More than 10 million • 25%
Top 10% • 25%
Top 1% • 25%
Top 5% • 25%
Below Top 10% • 25%