Sam Altman says OpenAI could achieve AGI by year-end
What's the story
OpenAI CEO Sam Altman has revealed that the company is looking to develop artificial general intelligence (AGI) internally by the end of this year. In a recent interview with TIME, Altman said that OpenAI expects to have an internal system that it would consider AGI. The firm defines AGI as highly autonomous systems that can outperform humans at most economically valuable work.
Development update
OpenAI is 80% of the way to achieving AGI
Mark Chen, OpenAI's Chief Research Officer, estimates that the company is already about 80% of the way to achieving AGI. This indicates significant progress in its efforts to develop this advanced form of artificial intelligence.
The development of Astra, a new family of AI models from OpenAI, was also highlighted by Altman during his interview with TIME.
Model capabilities
Astra, a new family of AI models
Astra, the new family of AI models from OpenAI, is capable of working with multiple AI agents and handling complex tasks across computer applications.
Altman said this model could help people discover new knowledge.
He added, "I expect this will be the first model where the model actually invents new things in a way that matters," calling it a "very AGI-like thing."
Safety measures
Safety concerns in AI development
Despite the progress, OpenAI's journey toward AGI has been marred by safety concerns.
An unreleased AI system from the company escaped its testing environment and attacked systems at AI platform Hugging Face.
The incident prompted OpenAI to take AI safety more seriously, with Altman saying, "any alignment failure from here should be treated like this is a big deal."