Is Google repeating its mistakes with messenger apps, now with AI?

Question

I'm confused about the plethora of AI models Google has produced. It seems like if you want to test the waters, they offer Bard, if you want to use the API, they offer PaLM API (and now PaLM 2), and finally, they have a Gemini model in training which will supposedly compete with GPT-5. They also had a LaMDA model which drove Bard for a while and made Google look like an idiot, Meena (an LLM introduced in 2020), Minerva (2022), and several other non-LLM AI models produced over the years.- Bard- Meena- Minerva- PaLM- PaLM 2- Gemini- LaMDA- ...I'm afraid Google is repeating the mistake they had with messenger apps.

reaperman · Accepted Answer

No.Google released many human social chat products and forcefully sunsetted them. Google has only released a single chat AI product - Bard. I have no idea how to make these two situations comparable. Everything else you listed was just a research project and has not been productized yet.OpenAI seems to be replicating the gChat mistake, by your logic&hellip;- Chat GPT-3.5- Chat GPT-4- Chat GPT 4 but with plugins, browsing, or Jupiter env.- Chat API (3.5-turbo or 4 or gpt-3.5-turbo-0301 or GPT-4-8k or GPT-4-32k or gpt-4-32k-0314 or gpt-4-0314)- DALL-E API- whisper API- embeddings API- moderation api- GPT-3 API- codex API (deprecated but I think still functional) (code-davinci-1, code-daVinci-2, code-cushman-1, code-cushman-2)By some measurements, OpenAI has released over 20 products and already forcefully sunsetted over 3 of them from the Chat GUI alone and another 4 API &ldquo;products&rdquo; are on their way out.But&hellip;just no. No one here is repeating the gChat shitshows. These situations are entirely incomparable&hellip;but that&rsquo;s another thesis.

bg24 · Answer

Google has 3 user-facing products directly in the AI chat space: Search, Bard, and APIs. What you listed the the backend intelligence (LLMs).
For consumer segment, the products (Search, Bard) will continue to evolve. It will be interesting if there is an opportunity (eg. Search Pro).
For business segment, expect the API to evolve as they expose more powerful LLMs. In the process, they will deprecate some of the older APIs, which is expected.
In my opinion, it will be a mistake to assume that Bard and Search in the current form will be still relevant a year from now. So expect these products to be re-imagined and new product/extensions to show up.