Friday, May 24, 2024

Weekly Review 24 May 2024

Some interesting links that I Tweeted about in the last week (I also post these on MastodonThreadsNewsmastBluesky and Post): 

  1. Another OpenAI safety team member quits due to a lack of confidence that the company's AI will ever be safe: https://futurism.com/openai-safety-worker-quit-confidence-agi
  2. The importance of AI to national security and prosperity: https://www.technologyreview.com/2024/05/13/1092322/why-america-needs-an-apollo-program-for-the-age-of-ai/
  3. Where are we going with the ethics of AI? https://www.informationweek.com/machine-learning-ai/designing-a-charter-for-navigating-the-evolving-ethics-of-ai
  4. Just because it's AI, doesn't mean it's appropriate: https://futurism.com/the-byte/students-graduation-speaker-ai-robot
  5. Unfortunately, politics and enormous egos arise in the AI industry as much as any other industry: https://futurism.com/the-byte/openai-leaks-musk-emails
  6. I don't believe that today's neural networks are conscious. Consciousness probably comes from quantum effects within neurons, and current AI have nothing like that: https://futurism.com/the-byte/ilya-sutskever-leaves-openai
  7. Is AI the Great Filter of technological civilizations? Are aliens just staying the hell away from us? https://futurism.com/the-byte/alien-civilizations-destroyed-by-ai
  8. The impact of AI on data engineering: https://www.datanami.com/2024/05/15/ai-impacting-data-engineering-faster-than-expected-dbt-labs-handy-says/
  9. An overview of the AI regulatory landscape, especially in Europe: https://www.kdnuggets.com/all-about-the-ai-regulatory-landscape
  10. I've always thought that the publisher Hindawi was a bit dodgy, now 19 of its journals have been wound-up by its owner after AI generated papers were published in them: https://www.theregister.com/2024/05/16/wiley_journals_ai/
  11. A framework for ethics and minimising bias in AI: https://dataconomy.com/2024/05/10/ethical-ai-tackling-bias-and-ensuring-fairness-in-machine-learning-algorithms/
  12. I've said for some time that the movie Her is the most accurate representation of AI in the near future. Now Sam Altman has confirmed that it was his inspiration: https://futurism.com/the-byte/sam-altman-chatgpt-her
  13. There is no way I would want my credit card data to be shared with any AI: https://www.computerworld.com/article/2110274/visa-leverages-ai-to-help-retailers-access-more-customer-data.html
  14. Of course the companies that create training data for AI are exploiting people. How else are they going to get the data without paying huge sums? https://www.linkedin.com/pulse/copy-authoritarian-workplace-culture-cathy-b-glenn-uhxcc/
  15. The deal between Reddit and OpenAI is done, future AI will be trained on Reddit contributors' data: https://www.theverge.com/2024/5/16/24158529/reddit-openai-chatgpt-api-access-advertising
  16. Are there any predatory AI journals out there? I've certainly received many emails asking me to submit papers in ecology to such journals: https://www.nature.com/articles/d41586-024-01437-2
  17. If they want to test how well their AI sarcasm detector works, they should try it in New Zealand. We're a very sarcastic people: https://www.theguardian.com/technology/article/2024/may/16/researchers-build-ai-driven-sarcasm-detector
  18. The potential impact of AI in mathematics: https://www.nature.com/articles/d41586-024-01413-w
  19. More AI tools for education from Google: https://www.theverge.com/2024/5/16/24158484/google-gemini-ai-workspace-education-school-privacy-accessibility-chromebook
  20. Overview of the new AI models from OpenAI and Google: https://www.datanami.com/2024/05/14/new-genai-models-on-tap-from-google-openai/
  21. An AI that softens the tone used by angry callers to call centres: https://www.businessinsider.com/ai-take-anger-out-customer-complaints-call-centers-softbank-japan-2024-5
  22. How to setup Ollama, an open source local chatbot: https://www.computerworld.com/article/2106043/private-ai-chatbots-windows-pc.html
  23. There are differences in the brain structures of men and women, and AI can detect them: https://dataconomy.com/2024/05/16/ai-sex-based-brain-difference/
  24. Do we really need an international regulatory body for AI? We can't even convince the major world powers to prosecute war crimes: https://dataconomy.com/2024/05/13/sam-altman-ai-regulation-agency-2024/

No comments:

Post a Comment

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