OpenAI plans to release two highly anticipated models. Orion, possibly the new GPT-5 model, is expected to be an advanced large language model (LLM), while Strawberry aims to improve AI reasoning and problem solving, particularly in the domain of mathematics.
Why are these projects important?
The Strawberry project (previously called Q* or Q-Star) is reportedly a secretive OpenAI initiative to improve AI reasoning and decision-making to achieve more generalized intelligence. OpenAI co-founder Ilya Sutskever’s concerns about its risks led to the brief ouster of CEO Sam Altman. Unlike Orion, which focuses on optimizing existing LLMs like GPT-4 by reducing computational costs and improving performance, Strawberry aims to boost AI’s cognitive capabilities, according to the company. The information and Reuters Agency. OpenAI could even integrate Strawberry into ChatGPT to improve reasoning.
If true, what impact will they have on the technological world?
For autonomous systems, such as self-driving cars or robots, Strawberry could improve safety and efficiency. Future iterations could focus on improving interpretability, making their decision-making processes transparent. Tech giants like Google and Meta could face increased competition as customers in the healthcare, finance, automotive, and education sectors that are increasingly reliant on AI adopt OpenAI’s newer, improved models. Smaller startups could also struggle to compete with the new products, affecting their market position and investment prospects.
How can we be sure that OpenAI is developing this?
New investors appear to be interested in investing in OpenAI, which, according to The Wall Street Journalis planning to raise funding in a round led by Thrive Capital that would value it at more than $100 billion. Apple and Nvidia are potential investors in this round. Microsoft has already invested more than $10 billion in OpenAI, fueling reports that OpenAI is powering its AI models.
But can AI models really reason?
AI has a hard time reasoning like humans, but in March, researchers at Stanford and Notbad AI indicated that their Quiet-STaR model could be trained to think before responding — a step toward learning AI models to reason. DeepMind’s proposed framework for classifying the capabilities and behavior of artificial general intelligence (AGI) models recognizes that “emergent” properties of an AI model could give it capabilities like reasoning that are not explicitly anticipated by the model’s developers.
Will ethical concerns increase?
Despite claims of safe AI practices, big tech companies face skepticism due to past data misuse, copyright and intellectual property (IP) violations. AI models with enhanced reasoning could encourage misuse, such as misinformation. Quiet-STaR researchers admit that “there are no safeguards against harmful or biased reasoning.” Sutskever, who proposed what is now Strawberry, launched Safe Superintelligence Inc., with the goal of improving AI capabilities “as quickly as possible while making sure our safety is always ahead.”
Disclaimer:
The information contained in this post is for general information purposes only. We make no representations or warranties of any kind, express or implied, about the completeness, accuracy, reliability, suitability or availability with respect to the website or the information, products, services, or related graphics contained on the post for any purpose.
We respect the intellectual property rights of content creators. If you are the owner of any material featured on our website and have concerns about its use, please contact us. We are committed to addressing any copyright issues promptly and will remove any material within 2 days of receiving a request from the rightful owner.