In a latest Reddit AMA (ask me something), OpenAI CEO Sam Altman, together with another prime OpenAI executives, dropped a variety of hints concerning the firm’s future, and what to anticipate from ChatGPT subsequent 12 months.
Firstly, when requested if there could be a ChatGPT-5 Altman doubled down on his “faux information” response on X not too long ago to an article speaking of its imminent launch: “Now we have some excellent releases coming later this 12 months! Nothing that we’re going to name GPT-5, although”, he replied. The phrasing “this 12 months” does point out {that a} new LLM may very well be launched by OpenAI very quickly although.
Brokers are coming
It has been a busy week for OpenAI, with the discharge of Superior Voice mode on the desktop app variations of ChatGPT and a brand new ChatGPT search, which even challenges Google. When requested concerning the worth of the not too long ago launched ChatGPT search in comparison with conventional search engines like google, like Google, Altman was enthusiastic: “For a lot of queries, I discover it to be a method sooner/simpler strategy to get the data I am searching for. I believe we’ll see this particularly for queries that require extra complicated analysis. I additionally sit up for a future the place a search question can dynamically render a customized internet web page in response!”
When requested concerning the subsequent replace to Dall-E 3, the picture generator that’s a part of ChatGPT, Altman was non committal: “The subsequent replace will likely be definitely worth the wait! However we do not have a launch plan but.” So it doesn’t appear like a lot will likely be taking place on the picture era entrance for some time.
Actual intelligence
One space the place ChatGPT is being challenged by its rivals is in AI that may carry out duties autonomously. We’ve seen Google make inroads into this space with its Jarvis AI. When requested if ChatGPT will be capable of carry out duties by itself, Altman replied “IMHO that is going to be an enormous theme in 2025”, which signifies the route OpenAI will likely be taking subsequent 12 months.
Referring once more to autonomous AI, when requested what the subsequent massive breakthrough for ChatGPT could be, Altman replied: “We may have higher and higher fashions, however i feel the factor that may really feel like the subsequent big breakthrough will likely be brokers”. Brokers are autonomous AI bots that may carry out duties for you.
Maybe essentially the most fascinating remark from Altman was about the way forward for AGI – synthetic normal intelligence. Seen by many because the ‘actual’ AI, that is a man-made intelligence mannequin that might rival and even exceed human intelligence. Altman has beforehand declared that we may have AGI inside “just a few thousand days“.
When requested by a Reddit consumer whether or not AGI is achievable with recognized {hardware} or it can take one thing fully completely different, Altman replied: “We imagine it’s achievable with present {hardware}.”
