shrikant

"Loophole"? The secret ingredient is crime...

I remember the inverse of that in Chennai -- scheduled running water for an hour in the morning and an hour in the evenings when we had to stockpile water in every container possible.

I call this (well, in my head) the Sam Altman Fellow Kids style.

It's clearly a deliberate affectation that they're putting on because other words are properly capitalise (COVID, PIP, I, his own name, etc.). The author's other posts are properly formatted so I assume this was a a stylistic choice they made to illustrate their point on it being stream-of-consciousness.

I just find it more annoying on tweets because they've clearly fought the phone keyboard to override autocaps just to sound down with their fellow kids.

As always with smallish tools/services like these whipped up with an LLM, I find there's more value in seeing the "source code", i.e. the prompt itself.

Turns out it's quite useful in this instance, and I just made a small couple of tweaks to get a summary of my own workflow.

FWIW, this is what I modified it to:

Help me share how I work with AI with my teammates.

Use our existing conversations and relevant local tools within my permissions to discover my agents, harnesses, skills, connections and working practices. Keep secrets, raw configuration and private project details out of the draft. Explain how my tools work together and develop a concrete workflow where the evidence supports it, rather than listing installed software. Draft from what you know without a questionnaire; ask one short question only if an essential gap prevents a useful draft. If local inspection is unavailable, use our conversation without claiming otherwise. Save a private draft, show me the exact preview, and wait for my explicit approval before publishing.

What harness do you use for Gemini? Antigravity?

Also, that user's last four (three of them in the last hour) submissions have all been similar "finding" reports from a Claude-generated mystery research group website. All which contain exclusively AI slop articles. Ugh.

Pretty sure that OP ("jakobgreenfeld") is the "founder". That user's last four submissions have all been similar "finding" reports from a Claude-generated mystery research group website.

Agreed, I thought the subject matter was interesting enough to try and labour through the tedious prose, but once I got to "Their scale is the point." I just had to stop and just skim the rest.

Yeah Ctrl/Cmd+F works just fine on Firefox as well.

Especially since all OP seems to do is churn these out and exclusively only post them? Ugh, these AI thought leaders...

Yeah when I worked at Deloitte Consulting, we had similar training that said something to the effect of "don't put something in an email that wouldn't feel comfortable as a newspaper headline".

I believe the bigger failure mode in that scenario is the fact that the site was storing user passwords in plaintext (either directly or effectively). Users tend to reuse passwords, so when there's a breach, more than just the user accounts for that site would be compromised.