OpenAI and Meta are about to launch new AI systems. These systems can think and make plans, moving closer to superhuman intelligence in machines.
This week, leaders from OpenAI and Meta indicated they are getting ready to release new versions of their big language models. These models are behind AI tools like ChatGPT.
Meta announced the launch of Llama 3 will start soon. At the same time, OpenAI, with Microsoft’s support, suggested their new model, probably GPT-5, is about to be released “soon”.
“We are hard at work in figuring out how to get these models not just to talk, but actually to reason, to plan . . . to have memory,” said Joelle Pineau, vice-president of AI research at Meta.
Brad Lightcap, the COO of OpenAI, shared with the Financial Times that the new GPT version would make advancements in dealing with difficult tasks, including reasoning.
“We’re going to start to see AI that can take on more complex tasks in a more sophisticated way,” he said. “I think we’re just starting to scratch the surface on the ability that these models have to reason.”
Lightcap mentioned that current AI systems excel at single, small tasks but remain quite limited in what they can do.
The updates from Meta and OpenAI are among many new big language models coming out this year from firms like Google, Anthropic, and Cohere.
Tech companies are in a hurry to make better generative AI. This software can make words, pictures, code, and videos that seem like they were made by people. The speed of advancements is increasing.
Being able to reason and make plans is key for reaching what experts in AI term “artificial general intelligence,” or the ability to think like a human. These abilities let chatbots and virtual aids carry out groups of tasks and foresee the results of what they do.
Meta intends to integrate its latest AI model into WhatsApp and Ray-Ban smart glasses. The company will launch Llama 3 in various sizes for different uses and devices in the next few months.
During an event in London, Chris Cox, the Chief Product Officer at Meta, mentioned that the cameras in Meta’s Ray-Ban glasses could help examine things like a malfunctioning coffee machine. An AI assistant, running on Llama 3, would then guide the user on how to repair it.
What we think?
I think the new AI from OpenAI and Meta will be really smart. They’re making AIs that can think and plan like us.
Soon, we might talk to AIs that can help with lots of things, even fixing a coffee machine! Meta’s going to use its AI in glasses and apps, which sounds cool. It’s exciting to see how these new AIs will change what we do every day.
