Sassbook AI Summarizer uses state-of-the-art AI to generate automatic summaries of text documents. The default mode is "Abstractive" which is equivalent to how humans summarize.
There is also the more traditional "Extractive" mode where key sentences from the original form the summary.
There are three categories of summary sizes as well.
In addition, for very short text, a highly abstractive "one-liner" may be produced (warning: wild!).
Sassbook only generates completions that help the writer to express themselves. It serves to augment human capabilities and bring automation at a higher level that word-level auto-complete.
There are far more legitimate and useful outcomes when put to good use.
The narrative that spells doom around these technologies are at best marketing gimmicks.
Agree, and this is in general an issue with lots of "deep learning" AI technology. The models are trained on huge amounts of text in the public domain and biases often leak into them.
This is an active research area and we will actively try to address such issues.
Apologies if it offended you or anyone else.
Thanks for the feedback; these are important issues that we want surfaced.
Note that the generations aren't expected to be factually correct, but rather provides a possible completion, which may be edited for originality and correctness.
Yes, the SaaS does rely on state-of-the-art NLG. I don't know why you think they are far better if we are using them as is. I am not offended at all; just curious where that big difference is.
Who says one can't compete against the best minds? Isn't that the premise of startups and progress?
If you are a fan of "CTRL", you should have liked ours a bit better.
Apps need to simplify and yet provide the desired experience for end users. From your feedback, it seems we need to try way harder.
BTW, you could just delete and retry and you might end up with a better completion. It is expected to create different generations on each attempt where possible.
Thanks for trying it out and providing valuable feedback!
Could you please try hitting the "Continue" button? Since the first generation was reasonable, you could just continue until the generation starts to veer off.
A new prompt is needed to express the next thought and so on. Typically, 1-3 continuations are possible with decent results in most cases.
In any case, this is extremely useful feedback that we will address.
Could you please not give up and try the above suggestion?
That wouldn't be something that could be considered "to the extent reasonably needed to operate and provide the Services to you".
There is no such intent anyway. I'll review it with the lawyer and revise it to something that doesn't scare people away!
I suppose this is applicable in general to content (as far as lawyers are concerned) and to protect us in case there is a need legal necessity for compliance purposes.
There is also the more traditional "Extractive" mode where key sentences from the original form the summary.
There are three categories of summary sizes as well.
In addition, for very short text, a highly abstractive "one-liner" may be produced (warning: wild!).
Thanks for trying and providing feedback!
Sassbook AI