This doesn’t ring true for me. I have ChatGPT write code for me that I could write myself. I’ve had ChatGPT even rewrite code for me to make it more legible. It’s pretty good at it, especially when it comes to more popular languages.
I tried asking it to write code to swap columns 1 and 3 of a CSV file, written in only x86 assembly. It refused, claiming that while it was theoretically possible, it would be stupid to do such a thing. It couldn't be persuaded...
This is Google, who have a monorepo, who are afraid of the legal risks to the monorepo. The copyright of the output of these tools is not yet determined, it's a risk. The advice not to use the tools in a risky way not only directly mitigates the risk but like an "employees must wash hands" sign in a restaurant restroom, it can be seen as a reasonable step that transfers some of the liability to the staff member. It's…
I suspect they have a model trained on their own proprietary data, and would prefer employees use that, rather than the bard model, trained on mixed data that they don't own.
This is Google, who have a monorepo, who are afraid of the legal risks to the monorepo. The copyright of the output of these tools is not yet determined, it's a risk. The advice not to use the tools in a risky way not only directly mitigates the risk but like an "employees must wash hands" sign in a restaurant restroom, it can be seen as a reasonable step that transfers some of the liability to the staff member. It's…
> This is Google, who have a monorepo, who are afraid of the legal risks to the monorepo.
Where are you getting any of that from the linked article?
It’s a very important question. There are of course two obvious readings/real concerns: (1) the untested IP issues concerning models trained on code from outside the firm, (2) the possibility of unreliable code. From my experience working with LLMs heavily since January, I would articulate my philosophy as “LLMs cannot be authors” — this is just to say, “authorship” is less about having actually created or having done the work, but being legally liable which is to say having skin in the game. Assuming for a second IP issues are not a concern, my rule for developers (or content creators) post-GPT: “you now have more responsibility for errors in your code than before, use whatever tools allow you to be most certain.” I hope that this is the world we are entering, not of generalized irresponsibility or irrelevance of humans in an automated world, but that we finally focus on what we decide really needs doing (thus of course that deciding itself) and taking responsibility that it be done as best as possible, however that may be.
I doubt the fact they use a monorepo has any impact on this thinking. A versioned sourced repository is no different from a file server. If some "ai copyright infringed" code made it into the monorepo it doesn't compromise the whole thing. There are still clear distinct projects and products. But exactly, this is a hygiene thing. Staff will still be using these tools anyway.
I agree that monorepo does not make a fundamental difference. Still it makes it easier for some undesirable things to happen. The tainted code can become a critical dependency. It can get copy-pasted elsewhere. An engineer could look at the code before writing their own (obviously not illegal per se but makes it harder to repeal bullshit claims later). An engineer can just write similar code and then have no way to p…
> It can get copy-pasted elsewhere.
I have no experience working with a monorepo but I've read Google has "an army" of people maintaining the monorepo. I would have imagined with enough support like this, you wouldn't need to copy paste code everywhere in a monorepo?
I'd appreciate any insight anyone can share on monorepo. I am fascinated by the idea but I've never had a chance to work with it and people who I've met who have either don't want to or can't talk about their experience with it.
As we all know, bard sucks really bad right now even compared to Gpt3. For those in the know, how far behind is Google from Open AI? how far behind is LLama?
Wow I can't use bard yet as I'm in the EU. But ChatGPT has never done something like this to me.
ChatGPT is ALWAYS doing stuff like this to me. Suggest code that doesn't compile, suggest SIMD instructions that don't exist, reiterate on incorrect answers with the exact same answer, etc. Granted I use ChatGPT 3.5.
> suggest SIMD instructions that don't exist
well that's one way to make code faster, just require the underlying hardware to compute things for you