Friday, September 1, 2023

Weekly Review 1 September 2023

Some interesting links that I Tweeted about in the last week:

  1. Seems more like a ramraid detection system than a ramraid prevention system. I'm not sure how an #AI could prevent a ramraid, unless it's a terminator:
  2. The only people that don't fail sometimes are the ones who don't do anything. Why don't academics talk about failure more? PhD students especially need to hear about it:
  3. Why Chatbots are so weird and wrong:
  4. Using random forests to predict chain extinctions:
  5. The paper for predicting chain extensions:
  6. Code generated by #AI is pretty rubbish now, but it will get better. How can human coders stay relevant?
  7. In New Zealand, we don't need to write essays to enter university. But in places that do use this system, #AI tools can be of help with drafting the essay, especially for underprivileged applicants:
  8. Crediting an #AI as a co-author of scripts is like crediting MS Word as a co-author - in both cases, the computer is a tool to assist an actual creative person, not the entity doing the creating:
  9. Even unconscious biases tend to make their way into #AI. It's a good thing that Google wants to make their teams working on this more diverse:
  10. How to tell if / when an #AI becomes conscious:
  11. No, #AI will not replace your job, but it will change the way you do it:
  12. So Hollywood has been using #AI to assist worker to generate effects for quite a while:
  13. IEEE's list of top programming languages for 2023 - Python is still at the top, in part due to its popularity in #AI:
  14. An overview of performance measures for #AI:
  15. At least businesses are talking about the ethics and privacy of large language model based #AI:
  16. Some interesting projects using generative #AI:
  17. Some things an enterprise needs to consider and prepare before trying to use #AI:

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.