Jmail maintainer and co-creator here. Very excited to see that someone finally made Jemini good!
Our development process has been interesting. Although just Riley and I first made Jmail, it's been really gratifying to see companies, journalists, and fellow developers like Diego rise to the occasion to make this entire suite of apps as high quality and extensive as possible.
embedding-shape | 2026-02-16 20:35 UTC
> Jmail maintainer and co-creator here.
Great, I've been looking to be able to contact you somewhere, hardly a better place :) Thanks for sharing the project btw, and for keeping jmail alive, been useful to dig into some stuff.
For that example, the first problem is that there is no link to the source file from the disclosed files, which is strange, because most other emails have that. Secondly, this almost seems to be a "sponsored" email or something, as it has an ad in the top right corner reading "Sponsored by Drop Site News", but clicking that just takes you to some general page, with no clear information how that's related to the fake/injected email. Also, what does "Verified by X" actually mean, did they verify the authenticity of that specific email?
There seems to be a bunch of people confused by those emails and wondering where they are coming from, because it's missing the source link like the others, so there doesn't seem to be a good way to verify those emails. Could you maybe share a bit about what's going on?
3rodents | 2026-02-16 20:41 UTC
That’s just a campaign email with “Pedophiles” as the recipient name, right? Anyone can sign an email address up to a mailing list. All of Trump’s campaign emails are these type of overly friendly weird junk.
lukeigel | 2026-02-16 20:54 UTC
Yeah, those are mailing lists that people signed him up for after he died, including with a joke name.
He was a very famous figure in August 2019, and normal people spammed his inbox with emails like the below
Recently he said he had a $50k hosting bill but that the CEO of Vercel offered to cover them and help them start some optimization.
llbbdd | 2026-02-17 07:11 UTC
I can think of some ways you could benefit the world without spending a cent
lenerdenator | 2026-02-17 14:07 UTC
I just wanted to say that you're doing really important work.
ddtaylor | 2026-02-17 14:52 UTC
You guys are doing awesome stuff. I was in the process of loading all of the documents into a vector database to query and then you dropped Jemini.
I wanted to reach out and avoid duplication of efforts, but I didn't know the best person to reach out to for this or if this is anything that can be collaborated on.
krainboltgreene | 2026-02-16 19:37 UTC
This feels like an insanely bad idea.
sperr11 | 2026-02-16 19:57 UTC
Care to elaborate?
spankalee | 2026-02-16 20:01 UTC
If it's using an LLM it'll make stuff up... about people and sex trafficking.
8note | 2026-02-16 20:29 UTC
for another angle - depending on the provider, theyre going to train on these queries and responses, and i dont want folks training an Epstein LLM, or accidentally putting Epstein behaviour into LLMs
SlightlyLeftPad | 2026-02-16 20:01 UTC
Interesting, I had the opposite feeling.
heisgone | 2026-02-16 20:11 UTC
Great! This, along with Webb, the other Epstein AI tool, are amazing effort to find documents. The way you link to files instead of just outputing an answer is amazing. Great works!
venusenvy47 | 2026-02-16 20:25 UTC
The Jamazon interface is really nice. I like how clicking on the order number takes you to the email. But I'm not going spend much time on there, after seeing his order for Lolita next to orders for kid toys.
Interesting. The majority of items seems to be books.
SauntSolaire | 2026-02-16 20:49 UTC
Along with the orders for kids school uniform skirts a bit later.. tough viewing in context.
Makes it unsettling when it then shows orders for books you've also read. Hard to swallow having anything in common with such a person.
I would have also assumed he got his clothes from much fancier places than Amazon, considering his wealth. Can't beat two day shipping I guess.
Altogether his Amazon orders are disturbing in places, but also somewhat humanizing, which is not what I expected.
ZpJuUuNaQ5 | 2026-02-17 13:22 UTC
>Makes it unsettling when it then shows orders for books you've also read.
>but also somewhat humanizing, which is not what I expected.
I am obviously not defending him or anything, but it always puzzles me how any of this confuses people. Whoever we may be, we are also humans, sharing common human traits and habits, and other people are just as real as we are. Every living moment, you unavoidably share at least some of the common beliefs, ideas, preferences, biases, habits, norms, values, goals, desires, fears, tastes, jokes, expectations, opinions and more with people you'd otherwise absolutely abhor.
Interesting stuff! I'm also slowly diving into the whole Epstein thing, mainly focused on a human-curated wiki (at https://corroborators.wiki, still WIP obviously) and I've had some contributors wanting to use AI for basically filling out details automatically. But I'm using LLMs daily for programming, so I know how much they get wrong.
So about that specific thing, how do you avoid this particular problems with LLMs getting basic things wrong and basically being a lottery if the answer is correct or not? It's the reason I've avoided (and encouraged others to avoid it to) using LLMs too heavily in the process of reading, understanding and summarizing documents, but clearly you've must have figured out some trick to this?
widdershins | 2026-02-16 21:23 UTC
Ask it to cite the document and page, and verify it yourself.
SLWW | 2026-02-16 21:14 UTC
It's passing the initial sniff test; this might be the first valuable use of AI I've seen in a good minute
dvrp | 2026-02-16 21:17 UTC
We're also skeptical of how AI is being used. Let us know if you have obvious horrible mistakes so we can fix them.
jeron | 2026-02-17 00:42 UTC
i know it's hyperbole but come on
pembrook | 2026-02-17 00:48 UTC
Yea every other use case beyond feeding the hysterical Epstein conspiracy machine has been totally worthless. All the consumers and companies paying real money for this tech must just be stupid.
On the other hand, providing more fuel for sweaty basement dwellers to invent fake stories about elite lizard people while trashing the civil liberties of thousands of random innocents contacted by this dude? I can think of no better use case.
/s
llbbdd | 2026-02-17 03:14 UTC
Upvoting for making me feel sane. I don't understand the level of attention paid to this shit, and it makes the people invested in it seem orders of magnitude more perverted than the financier himself is purported to be.
_345 | 2026-02-17 04:04 UTC
i can't wait for this epstein crap to end
monegator | 2026-02-17 09:50 UTC
>All the consumers and companies paying real money for this tech must just be stupid
yes?
dirasieb | 2026-02-17 15:14 UTC
cry harder
mentalgear | 2026-02-17 09:20 UTC
After 10 years in it, semantic search (where the human is knowledgable enough to verify) has been and still is the most useful field for LLMs.
hkt | 2026-02-16 21:19 UTC
Why on earth does he have CPAP pipes, an oxygen concentrator, and a pulse oximeter? This is dated 2019 so I assume he was having sleep apnea, but in my nightmares I assume he's keeping someone in a coma.
Xss3 | 2026-02-16 21:52 UTC
Thats an incredibly unsettling thought.
Slothrop99 | 2026-02-17 00:33 UTC
His brother said he had a CPAP machine in his jail cell. (EFTA00113460)
llbbdd | 2026-02-17 03:22 UTC
What a crazy subthread. You identified an obvious normal reason for him to have those things and then invented another reason, and then people agreed with you.
hugh-avherald | 2026-02-17 08:25 UTC
Epstein Scandal Update: Epstein sleeps nude in a oxygen tent which he believes gives him SEXUAL POWERS!
tehjoker | 2026-02-16 21:43 UTC
When LLMs first came out, my initial impulse, though I didn't pursue it, was that the summarization function was ideal for sorting through FOIA requests. So glad to see stuff like this come to fruition.
varjag | 2026-02-16 22:01 UTC
Fedorovism at home
hokkos | 2026-02-16 22:04 UTC
this ai is too much "aligned" to return anything of value, considering the content it has to look into and the questions it needs to answer.
dvrp | 2026-02-16 22:14 UTC
What do you mean?
As in, OpenAI, Anthropic, and Google's models won't follow instructions regarding forensics for this?
NamlchakKhandro | 2026-02-17 08:32 UTC
Who here is naive enough to think that this little loop hole hasn't been nicely tied off?
lukeigel | 2026-02-16 22:48 UTC
Bring your own key and use Claude. We found that it's most willing to run deep research queries here.
pillefitz | 2026-02-17 05:30 UTC
Which model is being used? Is there an unrestricted open weight model that could be used?
Hey Jeffrey
Comments
Our development process has been interesting. Although just Riley and I first made Jmail, it's been really gratifying to see companies, journalists, and fellow developers like Diego rise to the occasion to make this entire suite of apps as high quality and extensive as possible.
Great, I've been looking to be able to contact you somewhere, hardly a better place :) Thanks for sharing the project btw, and for keeping jmail alive, been useful to dig into some stuff.
However, there seems to be some "injected" or "fake" emails, that I cannot figure out why they're there in the first place. For example this one: https://jmail.world/thread/55b91b46ef1e4487bee131a8505e14a4?...
For that example, the first problem is that there is no link to the source file from the disclosed files, which is strange, because most other emails have that. Secondly, this almost seems to be a "sponsored" email or something, as it has an ad in the top right corner reading "Sponsored by Drop Site News", but clicking that just takes you to some general page, with no clear information how that's related to the fake/injected email. Also, what does "Verified by X" actually mean, did they verify the authenticity of that specific email?
There seems to be a bunch of people confused by those emails and wondering where they are coming from, because it's missing the source link like the others, so there doesn't seem to be a good way to verify those emails. Could you maybe share a bit about what's going on?
He was a very famous figure in August 2019, and normal people spammed his inbox with emails like the below
https://jmail.world/thread/4accfb5f3ed84656e9762740081a4579?...
These are all real emails! We can do a better job making this clear to the user.
I wanted to reach out and avoid duplication of efforts, but I didn't know the best person to reach out to for this or if this is anything that can be collaborated on.
https://jmail.world/jamazon
Makes it unsettling when it then shows orders for books you've also read. Hard to swallow having anything in common with such a person.
I would have also assumed he got his clothes from much fancier places than Amazon, considering his wealth. Can't beat two day shipping I guess.
Altogether his Amazon orders are disturbing in places, but also somewhat humanizing, which is not what I expected.
>but also somewhat humanizing, which is not what I expected.
I am obviously not defending him or anything, but it always puzzles me how any of this confuses people. Whoever we may be, we are also humans, sharing common human traits and habits, and other people are just as real as we are. Every living moment, you unavoidably share at least some of the common beliefs, ideas, preferences, biases, habits, norms, values, goals, desires, fears, tastes, jokes, expectations, opinions and more with people you'd otherwise absolutely abhor.
If you like interesting ecom UIs, shamelessly plugging my Show HN from today: https://news.ycombinator.com/item?id=47041288
So about that specific thing, how do you avoid this particular problems with LLMs getting basic things wrong and basically being a lottery if the answer is correct or not? It's the reason I've avoided (and encouraged others to avoid it to) using LLMs too heavily in the process of reading, understanding and summarizing documents, but clearly you've must have figured out some trick to this?
On the other hand, providing more fuel for sweaty basement dwellers to invent fake stories about elite lizard people while trashing the civil liberties of thousands of random innocents contacted by this dude? I can think of no better use case.
/s
yes?
As in, OpenAI, Anthropic, and Google's models won't follow instructions regarding forensics for this?