<![CDATA[My manager is going to open a team discussion about #AI / #LLM use for software development.]]>My manager is going to open a team discussion about / use for software development. Much of the company uses or and , while AFAIK I am the only one refusing to use these tools.

What are some recent studies I can reference about the harms of LLM use? I know there's stuff about deskilling, environmental harms, and code quality, but I won't have a chance to research specific studies until the end of my work day.

]]>https://board.circlewithadot.net/topic/0f9f8050-7f30-4de8-8ff8-80a6cac6ca62/my-manager-is-going-to-open-a-team-discussion-about-ai-llm-use-for-software-development.RSS for NodeThu, 14 May 2026 23:25:05 GMTThu, 14 May 2026 19:07:17 GMT60<![CDATA[Reply to My manager is going to open a team discussion about #AI / #LLM use for software development. on Thu, 14 May 2026 19:43:37 GMT]]>@shadow53

Link Preview Image
AI Assistance Reduces Persistence and Hurts Independent Performance

AI Assistance Reduces Persistence and Hurts Independent Performance

favicon

(ai-project-website.github.io)

]]>https://board.circlewithadot.net/post/https://mastodon.online/users/mastodonmigration/statuses/116574638390114969https://board.circlewithadot.net/post/https://mastodon.online/users/mastodonmigration/statuses/116574638390114969Thu, 14 May 2026 19:43:37 GMT<![CDATA[Reply to My manager is going to open a team discussion about #AI / #LLM use for software development. on Thu, 14 May 2026 19:43:03 GMT]]>@shadow53 https://www.anthropic.com/research/AI-assistance-coding-skills

]]>https://board.circlewithadot.net/post/https://social.treehouse.systems/users/fundamental/statuses/116574636203371979https://board.circlewithadot.net/post/https://social.treehouse.systems/users/fundamental/statuses/116574636203371979Thu, 14 May 2026 19:43:03 GMT<![CDATA[Reply to My manager is going to open a team discussion about #AI / #LLM use for software development. on Thu, 14 May 2026 19:36:03 GMT]]>@shadow53 cc: @baldur

He has https://needtoknow.fyi/ which might be handy.

]]>https://board.circlewithadot.net/post/https://floss.social/users/alcinnz/statuses/116574608640237106https://board.circlewithadot.net/post/https://floss.social/users/alcinnz/statuses/116574608640237106Thu, 14 May 2026 19:36:03 GMT<![CDATA[Reply to My manager is going to open a team discussion about #AI / #LLM use for software development. on Thu, 14 May 2026 19:26:11 GMT]]>@shadow53 Then, there is the cost risk - all of them make a big loss at the moment (https://www.wheresyoured.at/premium-ais-circular-psychosis/). This means you will in the very near future pay 100-1000x more money. On top of that you will spend even more money because the more code you generate the more token you have to spend in the future for AI to go through your code.
The cost risk is real: https://finance.yahoo.com/sectors/technology/articles/ubers-anthropic-ai-push-hits-223109852.html

]]>https://board.circlewithadot.net/post/https://social.anoxinon.de/ap/users/115989705188194340/statuses/116574569844151470https://board.circlewithadot.net/post/https://social.anoxinon.de/ap/users/115989705188194340/statuses/116574569844151470Thu, 14 May 2026 19:26:11 GMT<![CDATA[Reply to My manager is going to open a team discussion about #AI / #LLM use for software development. on Thu, 14 May 2026 19:20:41 GMT]]>@shadow53 I would ask to measure that it indeed brings improvement and have additional validation (unit testing, static code analysis, fuzzing etc.) that it does not decrease quality. Static code analyzers give you also statistics on duplicated code and code complexity. They can be indicator that things go wrong. Additionally, you can collect number of incidents (if LLM would work they should go down).

]]>https://board.circlewithadot.net/post/https://social.anoxinon.de/ap/users/115989705188194340/statuses/116574548262134082https://board.circlewithadot.net/post/https://social.anoxinon.de/ap/users/115989705188194340/statuses/116574548262134082Thu, 14 May 2026 19:20:41 GMT