Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
theptip
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
35 ms
·
721.
▲
by
theptip
2y ago
In the social media business, “integrity” generally means policing hate speech/abuse/bullying/…
722.
▲
by
theptip
2y ago
> CHIEF achieved nearly 94 percent accuracy in cancer detection and significantly outperformed current AI approaches across 15 datasets containing 11 cancer types I would have thought that performance vs. human level is the most interest
723.
▲
by
theptip
2y ago
Presumably that is exactly when their taxi service rolls out? While this has a dramatic rhetorical flourish, I don’t think it’s a good proxy. Even if it was safer, it would be an unnecessarily high burden to clear. You’d be effectively wri
724.
▲
by
theptip
2y ago
Not if you are reading your emails…
725.
▲
by
theptip
2y ago
Humans copy/paste from SO too. Does that prove humans can’t reason?
726.
▲
by
theptip
2y ago
LLM as a term is becoming quite broad; a multi-modal transformer-based model with function calling / ReAct finetuning still gets called an LLM, but this scaffolding might be all that’s needed. I’d be extremely surprised if AI recapitul
727.
▲
by
theptip
2y ago
Sometimes it is more narrowly scoped as “… economically valuable knowledge work”. But sure, if you have an un-embodied super-human AGI you should assume that it can figure out a super-human shelf-stocking robot shortly thereafter. We have A
728.
▲
by
theptip
2y ago
AKA “slicing the bologna”.
729.
▲
by
theptip
2y ago
Frankly, what you are describing is a money-printing machine. You should expect anyone who has figured out such a thing to keep it as a trade secret, until the FOSS community figures out and publishes something comparable. I don’t think the
730.
▲
by
theptip
2y ago
> The Scrum Guide would have you believe there are only two appropriate questions to ask a Product Owner A lot of people over-interpret rigid hierarchy’s, when Scrum is supposed to be the opposite. From the actual scrum guide: https:&#x
731.
▲
by
theptip
2y ago
Plenty of undesirable stuff can live in beer. Brettanomyces will make your beer taste phenolic. Lactic Acid Bacteria will sour it (Lactobacillus and Pediococcus are two common contaminants). Oxygenation will gradually make the beer taste li
732.
▲
by
theptip
2y ago
It’s a Public Benefit Corporation (sub-type of C-corp) not an LLC.
733.
▲
by
theptip
2y ago
> smooth out peak times not in the name of efficiency, but rather to lower the amount of dock workers the container terminal needs to hire. I’m confused. Efficiency means you don’t need to hire as much, since your peak-to-trough ratio is
734.
▲
by
theptip
2y ago
ChatGPT does this too by default right?
735.
▲
by
theptip
2y ago
You can turn GPUs into cash flow but the reverse isn’t guaranteed to be true.
736.
▲
by
theptip
2y ago
Compute is going to be king in the new world; if you have enough GPUs you can spam legal challenges. If you have more GPUs you can deploy those to take the other side and defend. As with DoS, the interesting cases will be where an asymmetry
737.
▲
by
theptip
2y ago
What data are you using to back up this belief? Progress on benchmarks continues to improve (see GPT-o1). The claim that there is nothing left to train on is objectively false. The big guys are building synthetic training sets, moving to mu
738.
▲
by
theptip
2y ago
The better reason to stop taking Altman at his word is on the subject of OpenAI building AGI “for the benefit of humanity”. Now that he’s restructuring the company to be a normal for-profit corp, with a handsome equity award for him, we sho
739.
▲
by
theptip
2y ago
This argument simply asserts that the LLMs (or their successor systems including scaffolding) will asymptote somewhere below human capabilities. It’s possible that this could happen but you need to propose a mechanism and metric for this ar
740.
▲
by
theptip
2y ago
Most people with theory of mind can’t trivially solve this problem though. So the test doesn’t disprove ToM in general, just that it memorizes some results.
741.
▲
by
theptip
2y ago
It takes minutes to find this specific issue in hindsight. It is laughable to claim that you can do a meaningful audit of a codebase in minutes.
742.
▲
by
theptip
2y ago
This is a great feature and I’ve been wishing for it for a long time. While the growth hacker dooming is asinine, one legitimate complaint might be that if I have to pick individual contacts I might not remember to select every friend I cou
743.
▲
by
theptip
2y ago
The simple way to visualize this is that all the processing can happen within a serializable transaction. If anything fails, the transaction rolls back. Marking the message as delivered is part of the same transaction.
744.
▲
by
theptip
2y ago
The alternative to consider is Meta RayBan. Personally I think that form factor is a more likely starting point as vision is going to be a big deal. But agreed that the pin (or ring / earring / other wearable) form factor is an ob
745.
▲
by
theptip
2y ago
If the danger is coming from the amount of compute invested, then cost of compute is irrelevant. A much better objection to static FLOP thresholds is that as data quality and algorithms increase, you can do a lot more with fewer FLOPs /
746.
▲
by
theptip
2y ago
How does it stack up vs. CLion?
747.
▲
by
theptip
2y ago
The underlying NotebookLM is doing better at this - each claim in the note cites a block of text in the source. So it’s engineered to be more factually grounded. I would not be surprised if the second pass to generate the podcast style lose
748.
▲
by
theptip
2y ago
If a malicious model exhilarates its weights to a Chinese datacenter, how do you turn that off? How do you turn off Llama-Omega if it turns out that it can be prompt-hacked into a malicious agent?
749.
▲
by
theptip
2y ago
This is true but the quality frontier is not a single bar. For mainstream content the bar is high. For super-niche content, I wouldn’t be surprised if NotebookLM already competes with the existing pods. This will be the dynamic of generated
750.
▲
by
theptip
2y ago
This is a great example. Areas where an expert can write a synthetic data generator (eg code, physics simulators, etc) are the dream scenario here. It seems to me there is a huge amount of unharvested low-hanging fruit here, for example IIU
More ›