When you use AI features, the IDE needs to send your requests and code to the LLM provider. In addition to the prompts you type, the IDE may send additional details, such as pieces of your code, file types, frameworks used, and any other information that may be necessary for providing context to the LLM.
Doesn’t sound like it gives you much transparency or control over the data it sends no matter which feature you use. Sadly not usable at my job then.
I worked for regulated companies too, I suspect that most companies would forbid such tools even if it’s not a dependency to build the software, especially if the source code is sent to random people. IMHO, every tool should be stored locally (e.g. on a proxy server) to make sure that the whole project can be recreated in a few minutes should something bad happens (or from scratch in a CI). As long as AIs rely on private companies on the internet, I wouldn’t use those tools.
They do say this though “We also plan to support local and on-premises models. For local models, the supported feature set will most likely be limited.”
This is currently a no go at my place (I asked) but the ai and security folks were interested in that as it would allow on-prem/private cloud usage as well as the possibility of using targeted models instead of a generic one.
For example, in the comments on their announcement they confirm they are looking at Azure AI support.
You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: !programming@programming.dev
Welcome to the main community in programming.dev! Feel free to post anything relating to programming here!
Cross posting is strongly encouraged in the instance. If you feel your post or another person’s post makes sense in another community cross post into it.
Hope you enjoy the instance!
Rules
Rules
Follow the programming.dev instance rules
Keep content related to programming in some way
If you’re posting long videos try to add in some form of tldr for those who don’t want to watch videos
Doesn’t sound like it gives you much transparency or control over the data it sends no matter which feature you use. Sadly not usable at my job then.
That’s a bummer. We’re strictly regulated and stuff like this needs to be self hosted or we can’t use it
I worked for regulated companies too, I suspect that most companies would forbid such tools even if it’s not a dependency to build the software, especially if the source code is sent to random people. IMHO, every tool should be stored locally (e.g. on a proxy server) to make sure that the whole project can be recreated in a few minutes should something bad happens (or from scratch in a CI). As long as AIs rely on private companies on the internet, I wouldn’t use those tools.
They do say this though “We also plan to support local and on-premises models. For local models, the supported feature set will most likely be limited.”
This is currently a no go at my place (I asked) but the ai and security folks were interested in that as it would allow on-prem/private cloud usage as well as the possibility of using targeted models instead of a generic one.
For example, in the comments on their announcement they confirm they are looking at Azure AI support.