Is Sam Altman just trying to get people to marvel at a few small strawberries he’s grown, or is the OpenAI CEO (the team behind ChatGPT) teasing the next big upgrade to its chatbot?
That’s what we’re left wondering after his latest social media post showing a plant pot growing fruit, complete with the caption “I love summer in the garden.” From anyone else, this would be a pretty basic even boring musing, but from Altman, many are taking this as a sign that OpenAI might soon launch its next big update for ChatGPT– which leaks from earlier this year said was codenamed ‘Strawberry’.
Currently, ChatGPT’s best commercially available model is GPT-4o. It rolled out back in May, and allowed the LLM to understand text, audio, image, and video prompts. It’s also faster than previous iterations, and smarter too. This was all showcased in an online demo in which ChatGPT was able to translate a conversation in real-time, and help someone to solve a math equation it was shown an image of among other neat tools.
Though, you might better know GPT-4o as the voice update which was meant to include a flirty Her-like voice called Sky before the specific personality was pulled – seemingly in response to legal action taken by actress Scarlett Johansson who claimed Sky was based on her, something OpenAI denied despite Altman tweeting out ‘Her’ the same day as GPT-4o’s launch (directly likening the voices to the movie starring Johansson as an AI).
GPT-5, or whatever official name ‘Strawberry’ takes, would supposedly improve the bot’s reasoning capabilities (per Reuters). The hope is this version of the AI could operate more independently from a human; not only could it simply find an answer to a question, but it could also do follow-up research – a major step towards a more human-like AI that other companies have yet to release.
Obviously, Altman’s post is nothing close to an official announcement. While it could be a teaser for Strawberry and hinting at an upcoming release, it could equally be Altman messing with people desperate to try Strawberry, or simply a post about his garden. That said, Altman only ever posts about AI and ChatGPT, so a sudden garden pic is more than a little out of character unless it means something more.
The only other evidence that points to this maybe being something more are posts on social media discussing an ‘Anonymous-chatbot’ that appeared in LMSYS Chatbot Arena. It allows people to see how two different models fare with the same prompt, and this new unknown model calls itself a fine-tuned version of GPT-4. Its appearance is being taken as a sign that some kind of announcement is on its way.
OpenAI is almost certainly cooking up an update to ChatGPT, it's just a question of whether or not Altman's post is explicitly referencing that next upgrade, and if it's coming soon. Signs certainly point to it but we wouldn't recommend holding your breath.
If you're desperate to see AI upgrades, your next best bet will be seeing what Google announces for Gemini at its Made By Google event next week.
Continue reading...
That’s what we’re left wondering after his latest social media post showing a plant pot growing fruit, complete with the caption “I love summer in the garden.” From anyone else, this would be a pretty basic even boring musing, but from Altman, many are taking this as a sign that OpenAI might soon launch its next big update for ChatGPT– which leaks from earlier this year said was codenamed ‘Strawberry’.
Currently, ChatGPT’s best commercially available model is GPT-4o. It rolled out back in May, and allowed the LLM to understand text, audio, image, and video prompts. It’s also faster than previous iterations, and smarter too. This was all showcased in an online demo in which ChatGPT was able to translate a conversation in real-time, and help someone to solve a math equation it was shown an image of among other neat tools.
Though, you might better know GPT-4o as the voice update which was meant to include a flirty Her-like voice called Sky before the specific personality was pulled – seemingly in response to legal action taken by actress Scarlett Johansson who claimed Sky was based on her, something OpenAI denied despite Altman tweeting out ‘Her’ the same day as GPT-4o’s launch (directly likening the voices to the movie starring Johansson as an AI).
GPT-5, or whatever official name ‘Strawberry’ takes, would supposedly improve the bot’s reasoning capabilities (per Reuters). The hope is this version of the AI could operate more independently from a human; not only could it simply find an answer to a question, but it could also do follow-up research – a major step towards a more human-like AI that other companies have yet to release.
Obviously, Altman’s post is nothing close to an official announcement. While it could be a teaser for Strawberry and hinting at an upcoming release, it could equally be Altman messing with people desperate to try Strawberry, or simply a post about his garden. That said, Altman only ever posts about AI and ChatGPT, so a sudden garden pic is more than a little out of character unless it means something more.
The only other evidence that points to this maybe being something more are posts on social media discussing an ‘Anonymous-chatbot’ that appeared in LMSYS Chatbot Arena. It allows people to see how two different models fare with the same prompt, and this new unknown model calls itself a fine-tuned version of GPT-4. Its appearance is being taken as a sign that some kind of announcement is on its way.
OpenAI is almost certainly cooking up an update to ChatGPT, it's just a question of whether or not Altman's post is explicitly referencing that next upgrade, and if it's coming soon. Signs certainly point to it but we wouldn't recommend holding your breath.
If you're desperate to see AI upgrades, your next best bet will be seeing what Google announces for Gemini at its Made By Google event next week.
You might also like
- OpenAI has a new scale for measuring how smart their AI models are becoming – which is not as comforting as it should be
- If you think GPT-4o is something, wait until you see GPT-5 – a 'significant leap forward'
- ChatGPT just (accidentally) shared all of its secret rules – here's what we learned
Continue reading...