So the govt implements rules and a panopticon for penalties. this works for the FDA, why wouldn't it for the FCC

Because regulation is bad, according to the current executive?

Politics aside, the FDA applies a very generous amount of regulation (mostly justifiable), not sure we want to pay multiples for our consumer electronics, as it (mostly) shows acceptable behavior and rearely kills anybody.

your anecdote doesn't have much to do with the reasons presented in the article?

> Writing code is the default behavior from pre-training

what does this even mean? could you expand on it

He means that it is heavily biased to write code, not remove, condense, refactor, etc. It wants to generate more stuff, not less.

Because there are not a lot of high quality examples of code edition on the training corpora other than maybe version control diffs.

Because editing/removing code requires that the model output tokens for tools calls to be intercepted by the coding agent.

Responses like the example below are not emergent behavior, they REQUIRE fine-tuning. Period.

  I need to fix this null pointer issue in the auth module.
  <|tool_call|>
  {"id": "call_abc123", "type": "function", "function": {"name": "edit_file",     "arguments": "{"path": "src/auth.py", "start_line": 12, "end_line": 14, "replacement": "def authenticate(user):\n    if user is None:\n        return   False\n    return verify(user.token)"}"}}
  <|end_tool_call|>

I'm not disagreeing with any of this. Feels kind of hostile.

I clicked reply on the wrong level. And even then, I assure you I am not being hostile. English is a second language to me.

I don't see why this would be the case.

Have you tried using a base model from HuggingFace? they can't even answer simple questions. You input a base, raw model the input

  What is the capital of the United States?

And there's a fucking big chance it will complete it as

  What is the capital of Canada? 

as much as there is a chance it could complete it with an essay about the early American republican history or a sociological essay questioning the idea of Capital cities.

Impressive, but not very useful. A good base model will complete your input with things that generally make sense, usually correct, but a lot of times completely different from what you intended it to generate. They are like a very smart dog, a genius dog that was not trained and most of the time refuses to obey.

So, even simple behaviors like acting as a party in a conversation as a chat bot is something that requires fine-tuning (the result of them being the *-instruct models you find in HuggingFace). In Machine Learning parlance, what we call supervised learning.

But in the case of ChatBOT behavior, the fine-tuning is not that much complex, because we already have a good idea of what conversations look like from our training corpora, we have already encoded a lot of this during the unsupervised learning phase.

Now, let's think about editing code, not simple generating it. Let's do a simple experiment. Go to your project and issue the following command.

  claude -p --output-format stream-json "your prompt here to do some change in your code" | jq -r 'select(.type == "assistant") | .message.content[]? | select(.type? == "text") | .text'

Pay attention to the incredible amount of tool use calls that the LLMs generates on its output, now, think as this a whole conversation, does it look to you even similar to something a model would find in its training corpora?

Editing existing code, deleting it, refactoring is a way more complex operation than just generating a new function or class, it requires for the model to read the existing code, generate a plan to identify what needs to be changed and deleted, generate output with the appropriate tool calls.

Sequences of token that simply lead to create new code have basically a lower entropy, are more probable, than complex sequences that lead to editing and refactoring existing code.

Thank you for this wonderful answer.

It’s because that’s what most resembles the bulk of the tasks it was being optimized for during pre-training.

During pre-training the model is learning next-token prediction, which is naturally additive. Even if you added DEL as a token it would still be quite hard to change the data so that it can be used in a mext-token prediction task Hope that helps

You are discounting his experience with your own, while your example is not remotely relevant to his flask game example.

It sounded like he was trying to one shot things when he mentioned he would ask it to fix problems with no luck. It's an approach I've tried before with similar results, so I was sharing an alternative that worked for me. Apologies if it came across as dismissive

Yeah at that point, just arguing semantics

they are not though

> expert in human language development and cognitive neuroscience, Gary is a futurist able to accurately predict the challenges and limitations of contemporary AI

I'm struggling to reconcile how these connect and he has been installed as Head of AI at Uber. Reeks of being a huckster

I didn't know the Uber bit, but googling:

>...held the position briefly after Uber acquired his company, Geometric Intelligence, in late 2016. However, Marcus stepped down from the directorship in March 2017,

which maybe fits your hypothesis.

something left out there with government though is incentive misaligned and hence corruption, which is smaller in a private scale (exists nonetheless)

The bigger blocker isn't the technology or the fleet. It's commercial viability and Luddite and populist politicians.

It's a joke