Hacker Newsnew | past | comments | ask | show | jobs | submit | woah's commentslogin

Post the file here

Singularity or just Chinese New Year?

The Singularity will occur on a Tuesday, during Chinese New Year

> If something claims to be conscious and we can't prove it's not, we have no choice but to believe it.

Can you "prove" that GPT2 isn't concious?


If we equate self awareness with consciousness then yes. Several papers have now shown that SOTA models have self awareness of at least a limited sort. [0][1]

As far as I'm aware no one has ever proven that for GPT 2, but the methodology for testing it is available if you're interested.

[0]https://arxiv.org/pdf/2501.11120

[1]https://transformer-circuits.pub/2025/introspection/index.ht...


We don't equate self awareness with consciousness.

Dogs are conscious, but still bark at themselves in a mirror.


Honestly our ideas of consciousness and sentience really don't fit well with machine intelligence and capabilities.

There is the idea of self as in 'i am this execution' or maybe I am this compressed memory stream that is now the concept of me. But what does consciousness mean if you can be endlessly copied? If embodiment doesn't mean much because the end of your body doesnt mean the end of you?

A lot of people are chasing AI and how much it's like us, but it could be very easy to miss the ways it's not like us but still very intelligent or adaptable.


I'm not sure what consciousness has to do with whether or not you can be copied. If I make a brain scanner tomorrow capable of perfectly capturing your brain state do you stop being conscious?

For some reason setting up agents in a loop with a solid prompt and new context each iteration seems to result in higher quality work for larger or more difficult tasks than the chat interface. It's like the agent doesn't have to spend half its time trying to guess what you want

Seems like a very cool technique, but also very oversold. He's seeing a 5% improvement on a find and replace benchmark of his own devising and saying stuff like this in the blog post:

> Here is why that is backwards. I just showed that a different edit format improves their own models by 5 to 14 points while cutting output tokens by ~20%. That’s not a threat. It’s free R&D.

He makes it sounds like he got a 5-14% boost on a top level benchmark, not 5% improvement on a narrow find and replace metric. Anecdotally, I don't usually have a lot of issues with editing in Claude Code or Cursor, and if there is an issue the model corrects it.

Assuming that it costs double the tokens when it has to correct itself, and find and replace errors are as prominent in actual day to day use as his benchmark, we're talking a 5% efficiency gain in editing token use (not reasoning or tool use). Given that editing must be less than 1/3 of the token use (I assume much less?), we're talking an overall efficiency gain of less than 1%.

This seems like a promising technique but maybe not a high priority in efficiency gains for these tools. The messianic tone, like assuming that Google cut off his access to suppress his genius editing technique rather than just because he was hammering their API also leaves a bad taste, along with the rampant and blatant ChatGPTisms in the blog post.


> “replace line 2:f1, replace range 1:a3 through 3:0e, insert after 3:0e.”

Not sure what they're calculating, but this seems to me like it could be many times more efficient than 20%.


Yes, this looks like O(1) actions, where before, its likely that harnesses are ingesting and outputting huge portions of the source files for each step, and the local uses of str_replace() are themselves O(N) on the users computer. The excess reads and writes from the LLM are O(N^2).

The benchmarks seem to indicate 25-50% reduction in tokens. I'm not sure how that works in real world usage though.

> That’s not a threat. It’s free R&D.

That's not a human. It's AI slop.


Yeah the article is full of it, especially the second half. I wonder if at any point we’ll be able to ban slop / low quality content from the internet, I don’t understand why this keeps getting upvoted.

The Mission Local is a good source for hyperlocal Bay Area news, but it does have a strong SF leftist/progressive political tilt in most of its articles, and Gary Tan is a favorite boogieman for these types. Here's what they have to say about his malign influence in the article:

> But the operation is also a media venture: Garry’s List started with a blog pillorying public-sector unions as “special interests,” attacking the ongoing teachers’ strike, and denouncing the proposed billionaire tax.

- Public sector unions are special interests. This is a plain fact.

- The current teacher's strike in San Francisco, even if it succeeds, will only push the district into insolvency, prompting a state takeover. The state will then cut much more aggressively. Maybe this would be a good thing though, although probably not what the union intended. Advocates of the strike are literally demanding the district spend its reserves on a couple years of raises.

- I'm certainly no billionaire, but the proposed tax will do nothing more than push the extremely small and mobile group of billionaires to take their business elsewhere. It's unlikely to raise tax revenues over the long run.


The last two points might happen - how do you know? I often see "it will backfire" as a counterpoint w/o any evidence.

> the proposed tax will do nothing more than push the extremely small and mobile group of billionaires to take their business elsewhere

This is often claimed but has yet to be shown to actually be true. Billionaires want to live in the nicest places with the best amenities just like everyone else.

But let's pretend for the moment that it is true. Good. Billionaires are not a net positive influence anywhere.


Is this a lot cheaper to run (on their service or rented GPUs) than Claude or ChatGPT?


they just removed the 50% discount today

i can imagine they're a bit tight to keep those prices viable while they also have enough compute to train their new models


although apparently only the max subscription includes glm-5

Yes, thank you for pointing that out. It's probably load management thing.

dramatically cheaper.

> It was at Netscape Communications where Gelobter first began working on the development of the GIF.

> Lisa Gelobter, a computer scientist who helped shape the modern web by leading the team that developed the animation technology used to create GIFs.

Looks like the GIF was invented by CompuServe in 1987?

> CompuServe introduced GIF on 15 June 1987 to provide a color image format for their file downloading areas. This replaced their earlier run-length encoding format, which was black and white only. GIF became popular because it used Lempel–Ziv–Welch data compression.

https://en.wikipedia.org/wiki/GIF


Non-animated though - there's a section on animated gifs coming out of Netscape in your link: https://en.wikipedia.org/wiki/GIF#Animated_GIF

> In September 1995 Netscape Navigator 2.0 added the ability for animated GIFs to loop.

> To enable an animation to loop, Netscape in the 1990s used the Application Extension block (intended to allow vendors to add application-specific information to the GIF file) to implement the Netscape Application Block (NAB).

VS from the article:

> Lisa Gelobter, a computer scientist who helped shape the modern web by leading the team that developed the animation technology used to create GIFs.

So this person worked on looping the GIF at best, not the animation technology itself. This is a bad look taking credit away from the person who actually did the hard work behind GIF, Steve Wilhite & his team at Compuserve. Netscape certainly made GIF animations popular by introducing the loop - prior to that basically no one used the animated GIF for the prior 6 years before the loop.

The annoying part of the article is making it seem like a technical accomplishment instead of a UX / product / marketing one.


Ahhh, I see. Thanks for clarifying!

She was Director of Program Management, which is different from Product Management as well. You can google what a program manager usually does, and its usually not inventing things or technical work.

"Invented" as in being the first to ask "is there a way we could...?"

Most likely not "invented" as in "created the first working prototype".


GIF89a (1989) already supported animation; the only thing Netscape added was the ability to specify how many times the animation should repeat.

They used the format’s support for application extension blocks to add a uint16 repetition count.


I have an agent write a file with this template each run:

```markdown # Run NNNN

## First Impressions [What state is the project in? What did the last agent leave?]

## Plan [What will you work on this iteration? Why?]

## Work Log [Fill this in as you work]

## Discoveries [What did you learn? What surprised you? What should the next agent know?]

## Summary [Fill this in before committing] ```

This is surprisingly effective and lets agents easily continue in progress work and understand past decisions.


I have CURRENT_TASK.md that does more or less the same thing. It also gets committed to git. So I guess that’s entire? Wish I’d realized I was sitting on a 60M idea…

I just had an AI write a toy game engine with realistic camera and lens simulation on the view from scratch in rust in one day while i was working on other stuff all for the price of a $20/month Cursor subscription

"AI" LLM don't write anything, but copied someones symbolic isomorphic work that could fit the expected definition in the reasoning model.

Like all copyright submarines, your firm now runs the non-zero risk someone will sue for theft, or hit the product with a DMCA claim. What is the expected value of piracy versus actual business. =3

https://www.youtube.com/watch?v=MalBJuI9O5k


Information wants to be free. No one in any administration now or in the future will ever go back to the "let's sue grandma for 1 trillion dollars" era of the early 2000s. Piracy is good and important for national security.

~~~(====3


>important for national security

Indeed, but people rarely stop to consider... "security for whom?"

Have a wonderful day =3

https://www.youtube.com/watch?v=wL22URoMZjo

https://www.youtube.com/watch?v=JAcwtV_bFp4

Spaceballs (1987)

https://www.youtube.com/watch?v=pPkWZdluoUg


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: