Computational Intelligence

Weekly Review 23 May 2025

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):Running a coffee maker with AI: https://arstechnica.com/gadgets/2025/05/the-tinkerers-who-opened-up-a-fancy-coffee-maker-to-ai-brewing/ But can it make a really good cup of tea?Why sales teams are struggling with AI: https://www.techrepublic.com/article/news-sales-industry-generative-ai-breakthrough/New positions are needed in organisational IT departments to deal with AI: https://www.theregister.com/2025/05/15/new_job_machine_learning_administrator/Meta continues to bend the rules while acquiring training data for its AI: https://arstechnica.com/tech-policy/2025/05/meta-is-making-users-who-opted-out-of-ai-training-opt-out-again-watchdog-says/Do! Not! Use! AI! To! Write! Legal! Filings! https://arstechnica.com/tech-policy/2025/05/judge-initially-fooled-by-fake-ai-citations-nearly-put-them-in-a-ruling/AI can learn to work together: https://www.theguardian.com/technology/2025/may/14/ai-can-spontaneously-develop-human-like-communication-study-findsI don't think a ban on AI regulation is going to be helpful: https://www.theverge.com/news/666288/republican-ai-state-regulation-ban-10-yearsOpen source is the future of AI: https://www.theregister.com/2025/05/14/future_of_llms_is_open/Generative AI can now design Lego constructions: https://arstechnica.com/ai/2025/05/new-ai-model-generates-buildable-lego-creations-from-text-descriptions/Reasoning AI models will soon slow their rate of improvement: https://techcrunch.com/2025/05/12/improvements-in-reasoning-ai-models-may-slow-down-soon-analysis-finds/As AI models advance, they are going to require more frequent safety testing: https://techcrunch.com/2025/05/14/openai-pledges-to-publish-ai-safety-test-results-more-often/ But will companies take the cheap route and use AI to do it?Compute bottlenecks will soon cause the rate of improvement of reasoning AI models to slow: https://www.techrepublic.com/article/news-ai-reasoning-models-scale/AI is getting entrenched in organisations before the security problems associated with it are understood: https://www.theregister.com/2025/05/14/cyberuk_ai_deployment_risks/AI generated material is increasing paranoia: https://arstechnica.com/ai/2025/05/welcome-to-the-age-of-paranoia-as-deepfakes-and-scams-abound/Google launches a new seed fund for AI startups: https://techcrunch.com/2025/05/12/google-launches-new-initiative-to-back-startups-building-ai/AI might or might not make good therapists, but using them is certainly strengthening government surveillance capabilities: https://www.theverge.com/policy/665685/ai-therapy-meta-chatbot-surveillance-risks-trumpAI can now create algorithms, not just write code to implement existing algorithms: https://spectrum.ieee.org/deepmind-alphaevolveHow AI is contributing to DevOps: https://www.datasciencecentral.com/the-future-of-devops-using-ai-automation-and-hpc/The challenges of AI in video surveillance: https://www.datasciencecentral.com/5-challenges-in-implementing-ai-in-video-surveillance/An IEEE standard on AI system procurement: https://spectrum.ieee.org/ieee-ai-3119-standardsA list of seven AWS services that are useful for AI: https://www.kdnuggets.com/7-aws-services-for-machine-learning-projectsA list of ten free books on AI: https://www.kdnuggets.com/10-free-artificial-intelligence-books-for-2025Open source is the future of AI: https://www.theregister.com/2025/05/14/future_of_llms_is_open/William Gibson's Sprawl trilogy had "Personality Constructs" that could recreate a person's thoughts and personality. This might soon be possible using AI: https://techcrunch.com/2025/05/15/sam-altmans-goal-for-chatgpt-to-remember-your-whole-life-is-both-exciting-and-disturbing/

May 23, 2025 - 13:20
 0
Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):
  1. Running a coffee maker with AI: https://arstechnica.com/gadgets/2025/05/the-tinkerers-who-opened-up-a-fancy-coffee-maker-to-ai-brewing/ But can it make a really good cup of tea?
  2. Why sales teams are struggling with AI: https://www.techrepublic.com/article/news-sales-industry-generative-ai-breakthrough/
  3. New positions are needed in organisational IT departments to deal with AI: https://www.theregister.com/2025/05/15/new_job_machine_learning_administrator/
  4. Meta continues to bend the rules while acquiring training data for its AI: https://arstechnica.com/tech-policy/2025/05/meta-is-making-users-who-opted-out-of-ai-training-opt-out-again-watchdog-says/
  5. Do! Not! Use! AI! To! Write! Legal! Filings! https://arstechnica.com/tech-policy/2025/05/judge-initially-fooled-by-fake-ai-citations-nearly-put-them-in-a-ruling/
  6. AI can learn to work together: https://www.theguardian.com/technology/2025/may/14/ai-can-spontaneously-develop-human-like-communication-study-finds
  7. I don't think a ban on AI regulation is going to be helpful: https://www.theverge.com/news/666288/republican-ai-state-regulation-ban-10-years
  8. Open source is the future of AI: https://www.theregister.com/2025/05/14/future_of_llms_is_open/
  9. Generative AI can now design Lego constructions: https://arstechnica.com/ai/2025/05/new-ai-model-generates-buildable-lego-creations-from-text-descriptions/
  10. Reasoning AI models will soon slow their rate of improvement: https://techcrunch.com/2025/05/12/improvements-in-reasoning-ai-models-may-slow-down-soon-analysis-finds/
  11. As AI models advance, they are going to require more frequent safety testing: https://techcrunch.com/2025/05/14/openai-pledges-to-publish-ai-safety-test-results-more-often/ But will companies take the cheap route and use AI to do it?
  12. Compute bottlenecks will soon cause the rate of improvement of reasoning AI models to slow: https://www.techrepublic.com/article/news-ai-reasoning-models-scale/
  13. AI is getting entrenched in organisations before the security problems associated with it are understood: https://www.theregister.com/2025/05/14/cyberuk_ai_deployment_risks/
  14. AI generated material is increasing paranoia: https://arstechnica.com/ai/2025/05/welcome-to-the-age-of-paranoia-as-deepfakes-and-scams-abound/
  15. Google launches a new seed fund for AI startups: https://techcrunch.com/2025/05/12/google-launches-new-initiative-to-back-startups-building-ai/
  16. AI might or might not make good therapists, but using them is certainly strengthening government surveillance capabilities: https://www.theverge.com/policy/665685/ai-therapy-meta-chatbot-surveillance-risks-trump
  17. AI can now create algorithms, not just write code to implement existing algorithms: https://spectrum.ieee.org/deepmind-alphaevolve
  18. How AI is contributing to DevOps: https://www.datasciencecentral.com/the-future-of-devops-using-ai-automation-and-hpc/
  19. The challenges of AI in video surveillance: https://www.datasciencecentral.com/5-challenges-in-implementing-ai-in-video-surveillance/
  20. An IEEE standard on AI system procurement: https://spectrum.ieee.org/ieee-ai-3119-standards
  21. A list of seven AWS services that are useful for AI: https://www.kdnuggets.com/7-aws-services-for-machine-learning-projects
  22. A list of ten free books on AI: https://www.kdnuggets.com/10-free-artificial-intelligence-books-for-2025
  23. Open source is the future of AI: https://www.theregister.com/2025/05/14/future_of_llms_is_open/
  24. William Gibson's Sprawl trilogy had "Personality Constructs" that could recreate a person's thoughts and personality. This might soon be possible using AI: https://techcrunch.com/2025/05/15/sam-altmans-goal-for-chatgpt-to-remember-your-whole-life-is-both-exciting-and-disturbing/

Read More

Tags:

Previous Article

Proper Regulation Essential for AI Advancements

Next Article

FFO: CLM Is Dead?, AI Change Management, Litigaze + 24-Hr LIC Tickets

Related Posts

Weekly Review 18 April 2025

May 12, 2025  0

Weekly Review 9 May 2025

May 16, 2025  0

Weekly Review 16 May 2025

May 16, 2025  0