AJ007

5 ms·

1.
by AJ007 2mo ago
What are we talking about here? I didn't read the full article but I looked at the synopsis at the top: "Striped patterns, flickering lights, bright glare, and crowded visual environments such as supermarkets" With the exclus
2.
by AJ007 3mo ago
The real question is when do you transition from building it with codex/CC to the harness itself.
3.
by AJ007 3mo ago
RIP all of the open source European software initiatives.
4.
by AJ007 3mo ago
Your world sure makes impersonation based cyber crime a lot simpler.
5.
by AJ007 3mo ago
Little appreciated fact is news orgs have full time employees just dealing with licensing all day long, and they pay out millions of dollars when someone fucks up.
6.
by AJ007 3mo ago
I wonder how long before it's common knowledge that a LLM has no segregation of a user's instructions and any other text it reads?
7.
by AJ007 5mo ago
Not sure how your last point matters if 27b can run on consumer hardware, besides being hosted by any company which the user could certainly trust more than anthropic. OpenAI & Anthropic are just lying to everyone right now because if t
8.
by AJ007 5mo ago
"end-to-end message encryption is a sham as long as" -- I agree with that but would add even more caveats. If someone can't list those caveats off the top of their head they shouldn't be pretending they aren't able
9.
by AJ007 5mo ago
Every person and company I know who had an Adsense account was banned and not paid. Two of them were banned for terms violations which were things Google reps told them to do. Endless conspiracy theories on this, no idea.
10.
by AJ007 5mo ago
Are you a lobbyist for Google, Apple, Meta, or the adtech industry? Because if you aren't, you are parroting their bullshit.
11.
by AJ007 6mo ago
A dirty secret is the algorithms can't differentiate real users from fake. The universe of content is so large now, if you don't start with a fake audience you go nowhere. Slop rises to the top, because slopfarms can spend all th
12.
by AJ007 6mo ago
Discord & Roblox - no encryption, privacy, or anonymity on either of those platforms, by the way.
13.
by AJ007 6mo ago
How does that even begin to make sense? I want to protect my child from X type of content -- one of many jobs of a parent, but I will trust all content to self report to be child inappropriate? "Inappropriate" is entirely subjecti
14.
by AJ007 6mo ago
Very much the puff piece of someone living in a social media bubble. The real problem is how the fediverse is going to survive the onslaught of laws related to social media age verification, data retention, data privacy, data not-privacy (b
15.
by AJ007 7mo ago
This whole thing is silly, LLMs can automate reference validation. If someone is a lawyer, accountant, doctor, teacher, surgeon, engineer etc, and is regurgitating answers that were pumped out with with GPT-5-extra-low or whatever mediocre
16.
by AJ007 7mo ago
Generative AI raises a lot of questions as to the value of copyright to society. There's a very dangerous direction I suspect things are tipping toward with generative AI: the big creative rights holders / representatives are goin
17.
by AJ007 7mo ago
You'll be arrested for some weird law that doesn't make sense, but it's ok because a pool of 12 people off the street won't consider whatever random thing you did a real crime!
18.
by AJ007 7mo ago
Every stage of human history was transitory, here comes the new one.
19.
by AJ007 7mo ago
When I was a kid, they said don't meet strangers you talked to online. That was it. Sometimes it turned out poorly, but as it could anywhere. Perhaps on the internet it was less risky because the person you were talking to had no idea
20.
by AJ007 7mo ago
Thinking about this a little more -> "nature of LLMs and intelligence" Bloated apps are a material disadvantage. If I'm in a competitive industry that slow down alone can mean failure. The only thing Claude Code has going
21.
by AJ007 7mo ago
OpenAI, Anthropic, Google, Microsoft certainly desire path dependence but the very nature of LLMs and intelligence itself might make that hard unless they can develop models which truly are differentiated (and better) from the rest. The Chi
22.
by AJ007 7mo ago
The British press has strict libel laws, and stories that the American press publishes are absolutely not legal in Britain, yet the British state doesn't threaten the New York Times or Washington Post with reparations (unless I'm
23.
by AJ007 7mo ago
I'm curious about the Slack thing. I wonder if there could be third parties doing something (browser plugins, third party keyboards for Android, edit: someone using a TV as a computer monitor.) One thing is for certain, if ad targeting
24.
by AJ007 7mo ago
Maybe the whole thing was intentional, right at the footer of viva "Cloud services by Microsoft Azure" ; #1 I've never heard of viva before #2 I've never seen an azure logo at the footer of a website.
25.
by AJ007 7mo ago
Extensions are trivial unless they have to run external software or services. Download the extension, extract the source, audit it with a good thinking model and either strip out all third party URLs/addresses or have the agent clone t
26.
by AJ007 7mo ago
They need to have an always-on camera looking at the person using the device. No camera, no discord.
27.
by AJ007 7mo ago
It is only a matter of time before ID verification means the camera is always on watching the face of the person looking at the screen.
28.
by AJ007 8mo ago
There is a good parallel here with Myspace and Facebook. Myspace added an ad network & was hammered by spammers around the same time Facebook was opening up user registration to everyone. Facebook had no ads. Myspace was dead. This time
29.
by AJ007 8mo ago
The YT transcripts are linked to on the YT page itself. If they remove that, it is trivial to use a local STT model to transcribe the video. If they make it impossible to download a video, you could just have a microphone record all of the
30.
by AJ007 8mo ago
One thing that's consistent with AI negative/doesn't work/is slop posts: they don't tell you want models they are using.

More ›