I received #GoogleBard access almost immediately after requesting it on opening day. I have been using it for a few hours and have some initial observations.
1. One of the biggest limitations is the number of input characters. It is under 4,000 from my informal testing. This means it is less useful for things like analyzing, summarizing, and adapting text.
2. It is slower than Chat GPT 3.5, but faster than Chat GPT 4.
3. It attempts to cite sources (or perhaps tries to provide other places to read?), but, shows poor judgment. I asked it to write a short causes of WWI handout for students, and one of the cited sources looks like a scam ebook site, one is far from an authority, one from a
medium quality source, plus the Wikipedia. Bing does a much better job at this point at citing sources and providing additional references.
4. I LOVE that Bard will show you other "drafts" it considers before showing you the final answer. It gives you a sense of what it might be thinking.
I hope this feature sticks around past this early beta.
5. It does all of the fun silly things... like my favorite silly prompt... asking it to write a Seinfeld something (episode, treatment, plot summary, dialogue).
Overall, it feels like earlier iterations of ChatGPT. I'd imagine that with user feedback, it will evolve quickly, much like OpenAI's models.
Share this Scrolly Tale with your friends.
A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.