All of there IoT devices will be slop coded soon, and I wonder whether that will be an improvement or not. I bet that security will be better.
All of there IoT devices will be slop coded soon, and I wonder whether that will be an improvement or not. I bet that security will be better.
> I bet that security will be better.
Not doxing myself, but... Company with a known name vibecoded a dashboard with Claude. Which also hardcoded a password into the client-side of the dashboard, which I caught.
I reckon security will be about the same.
> Not doxing myself
Seems irrelevant to the comment to add this, James. It just screams to do it a-la Streisand effect..
When I'm reading reviews of plans created by an agent especially on security boundaries it's suggesting huge matrixes to test even the very obscure situations, but then I'm also reading things like this and I just don't understand. Are we even using the same tools?
A large part is also how much you read back what the model writes. The good models generally write quite secure code, but they also often implement temporary solutions that they tell you to fix later.
For example, if secret storage methods aren't specified in the prompts, a model might decide to be clever and implement a generic secret access interface, with a default implementation that hardcodes everything. It will probably tell you that this is not production ready and you should write or specify your preferred secret storage implementation, but if you don't read or understand that, you'll just leave it as is and push to prod.
Management think models mean juniors can do senior work. Juniors don't know the footguns. Juniors can't read the code that the system outputs. Models get overwhelmed in any decent sized codebase.
Why would you be surprised there are failures?
[flagged]
Prod bugs up 260% since AI approval? Yeah, I'm gonna disagree.
BAs using AI make worse designs, management makes seniors redundant making code worse, and then QA is also laid off to make room for agentic testing. The results are not a surprise in anyway.
Refusal to consider evidence to the contrary is certainly symptomatic of something, and it isn't a well-reasoned argument. "Us" sounds like the corporate leadership I work with that have actively pushed teams to incur mountains to tech debt. I was walking over org metrics this week and sure enough, there's a rising tide of unplanned work for bugs, performance problems, and security snafus introduced during an aggressive push into GenAI with pressure to "just make agents do it."
All of which was, of course, predicted by management, reported upwards, and ignored.
1. It depends on model and tokens spent 2. Models talk the talk but not always walk the walk
Tools are already preventing IoT companies from doing a ton of things they do, by default. It's a problem of the process, churn, and culture, not tools. I don't have any doubts that if given a coding agent that cares more than they do they'd still force it to hardcode a password or something because they feel like it's more convenient. Nobody cares there.
> All of there IoT devices will be slop coded soon
Soon?
I've already seen multiple of TP-Link's firmware engineers leave their LLM history public and indexed by search engines.
It's quite obviously them as well.
How did you find that?