somewhereoutth

A good QA person/team will (be strongly incentivized to) make it their mission to find bugs and usability issues. Likely they will be domain experts, if only in the domain of the usage of your product. Worth their weight in gold, no machine can give you that.

Perhaps cheaper than upsetting your customers when they are the ones doing the testing?

No machine testing can completely (or even mostly) validate a human interface.

My feeling on testing:

- If it is used by a machine, then it can be tested by a machine.

- If it is used by a human, then it must be tested by a human.

Unfortunately it seems that nobody 'dialed back' LLM usage for the summary on that page - a good example of how un-human such text can feel to read.

My point is that many people do not have the time and opportunity.

If Men have the time and opportunity to do nothing but chill, or to have cool hobbies, then they are very definitely catching a break.

> We used to associate intelligence with accumulation. The smartest people were the ones who knew the most.

Anyone who was actually intelligent never saw any connection.

I don't see that - I do see failed projects, negative returns, and general disillusionment.

The only thing propping it up is the sunk cost fallacy, on now a biblical scale, requiring ever more hype to cover the rising disappointment (ChatGPT 5 when??). The blowback from the inevitable collapse will be similarly biblical in proportions, a wise strategy would be to avoid it all (as an individual and/or collectively at some level).

In fairness, there are / will be some areas where LLMs have business benefit, but likely those areas will be limited, and the 'AI stuff' will be hidden from the user.

> not doing [AI], either on an individual or a collective level, is just not an option.

I would have liked a deeper exploration of this point, since not doing it does indeed fix all the issues raised.

> for his own creations.

for their own creations is grammatically valid, and would avoid accusations of sexism!

So one houses humans nicely, the other houses humans horribly - and you support the horrible option?

> These people are born rich and die rich and affect very little positive change in the time between.

Sadly, with increasing wealth inequality, the rich are better able to keep out the talented but poor, with things like unpaid internships, access to professional networks gated by exorbitant college fees, insane rents in key cities, etc. There was a period in the mid 20 century when just talent and drive could get you very far. Should be no surprise that that led to a blossoming in the creative fields, and the converse more lately has led to their impoverishment.

The US is foundationally a European camping trip that got completely out of hand.

Heaps of fun, but eventually you get tired of things like public toilet ['restroom'] doors and the Rule of Law not working properly, so time to come home.

> As we learn to use LLMs in our work, we have to figure out how to live with this non-determinism. This change is dramatic, and rather excites me. I'm sure I'll be sad at some things we'll lose, but there will also things we'll gain that few of us understand yet. This evolution in non-determinism is unprecedented in the history of our profession.

The whole point of computers is that they were deterministic, such that any effective method can be automated - leaving humans to do the non-deterministic (and hopefully more fun) stuff.

Why do we want to break this up-to-now hugely successful symbiosis?

Because it pertains to the quality of the output - I can't validate every line of code, or test every edge case. So if I need a certain level of quality, I have to verify the process of producing it.

This is standard for any activity where accuracy / safety is paramount - you validate the process. Hence things like maintenance logs for airplanes.

Because when people use LLMs, they are getting the tool to do the work for them, not using the tool to do the work. LLMs are not calculators, nor are they the internet.

A good rule of thumb is to simply reject any work that has had involvement of an LLM, and ignore any communication written by an LLM (even for EFL speakers, I'd much rather have your "bad" English than whatever ChatGPT says for you).

I suspect that as the serious problems with LLMs become ever more apparent, this will become standard policy across the board. Certainly I hope so.

My original reframing was: the only way of showing that artificial intelligence can be constructed is by showing that humans cannot compute more than the Turing computable.

I was assuming the word 'compute' to have broader meaning than Turing computable - otherwise that statement is a tautology of course.

I pointed out that Turing computable functions are a (vanishingly) small subset of all possible functions - of which some may be 'computable' outside of Turing machines even if they are not Turing computable.

An example might be the three-body problem, which has no general closed-form solution, meaning there is no equation that always solves it. However our solar system seems to be computing the positions of the planets just fine.

Could it be that human sapience exists largely or wholly in that space beyond Turing computability? (by Church-Turing thesis the same as computable by effective method, as you point out). In which case your AGI project as currently conceived is doomed.

Humans do a lot of stuff that is hard to 'functionalise', computable or otherwise, so I'd say the burden of proof is on you. What's the function for creating a work of art? Or driving a car?

Function != Computable Function / general recursive function.

That's my point - computable functions are a [vanishingly] small subset of all functions.

For example (and close to our hearts!), the Halting Problem. There is a function from valid programs to halt/not-halt. This is clearly a function, as it has a well defined domain and co-domain, and produces the same output for the same input. However it is not computable!

For sure a finite alphabet can describe an infinity as you show - but not all infinity. For example almost all Real numbers cannot be defined/described with a finite string in a finite alphabet (they can of course be defined with countably infinite strings in a finite alphabet).

> then the only way of showing that an artificial intelligence can not theoretically be constructed to at least meet the same bar is by showing that humans can compute more than the Turing computable.

I would reframe: the only way of showing that artificial intelligence can be constructed is by showing that humans cannot compute more than the Turing computable.

Given that Turing computable functions are a vanishingly small subset of all functions, I would posit that that is a rather large hurdle to meet. Turing machines (and equivalents) are predicated on a finite alphabet / state space, which seems woefully inadequate to fully describe our clearly infinitary reality.

dispense with the parentheses:

  (eq (mul (a a)) (pow (a 2)))

becomes

  eq
    mul
      a a
    pow
      a 2

After factoring in the depreciation of their assets(the car)? Likely not.

It is a classic case of 'I need to eat today, so I'm taking from my future in order to do so' - this is how the poor get poorer (and the rich get richer).

My thinking tool, along with a ream of 80gsm blank white printer paper.

As well as being ubiquitous, reliable and cheap, you can also vary the line weight it produces with pressure. This makes it great for sketches and diagrams, as well as straight writing.

It is the blueprints (the detailed design, plans sections etc) that is analogous to code, not bricks. Software designers (compared to building designers) are lucky that the process of turning design (code) into artifact (running software) is virtually free in terms of cost and time. However software designers are unlucky that what they do is so misunderstood - not least by them themselves.

The crazy thing about ridesharing and similar, is that the workers are providing the capital (the cars), but the company is making the profits!

> But is computation enough?

Of course not, but that would invalidate the entire project of some/many here to turn reality into something clockwork they can believe they understand. Reality is much more interesting than that.

Knitting was a hipster thing in London (and elsewhere most likely) for about 6 months back in the dark days of peak hipsterism (2000-2010?). It takes rather a lot of time and effort, so I guess they got bored of it pretty quick.

"Anyone with access to AI tools can now produce work that, _at least superficially_, resembles professional output."

Key quote, emphasis mine.

Indeed. However the problem with LLMs is that vast amounts of VC money are being thrown at them, in the [misplaced] hope of great returns. This results in a resource mis-allocation of biblical proportions, of which unnecessary carbon emissions are a part.

Regardless of the mechanism, the foundational 'conceit' of LLMs is that by dumping enough syntax (and only syntax) into a sufficiently complex system, the semantics can be induced to emerge.

Quite a stretch, in my opinion (cf. Plato's Cave).