OpenAI and Meta are poised to unveil groundbreaking artificial intelligence models capable of reasoning and planning, marking significant strides towards achieving superhuman cognitive abilities in machines.
Executives from both OpenAI and Meta have indicated their plans to launch the next iterations of their large language models, which power generative AI applications like ChatGPT.
Meta announced the imminent rollout of Llama 3 in the coming weeks, while OpenAI, backed by Microsoft, hinted at the impending release of its next model, anticipated to be named GPT-5, in the near future.
Joelle Pineau, Meta's vice-president of AI research, emphasized the ongoing efforts to equip these models not just with conversational abilities but also with reasoning, planning, and memory capabilities.
Brad Lightcap, OpenAI’s chief operating officer, highlighted the upcoming advancements in the next generation of GPT, particularly in addressing challenging problems like reasoning.
While today’s AI systems excel in isolated tasks, they remain limited in their breadth of capabilities, as noted by Lightcap.
The upgrades from Meta and OpenAI form part of a broader trend of releasing new large language models this year by companies such as Google, Anthropic, and Cohere.
As tech companies strive to develop increasingly sophisticated generative AI, the pace of innovation is accelerating, with a focus on creating software capable of producing human-like outputs across various domains.
Reasoning and planning are crucial milestones on the path towards artificial general intelligence, enabling chatbots and virtual assistants to execute sequences of tasks and anticipate outcomes effectively.