Sign in

Justin M

@justinm.one
928 followers 606 following 1.2K posts

Working on security in tech. Opinions are my own :) If you have a large audience, please do not RT me

PostsRepliesMedia
Justin M @justinm.one · 1h
Strong +1 on this
010
Justin M @justinm.one · 7h
It sounds an awful lot like optimizing for engagement is the only way to retain users, and IMO this practice should be outright illegal for any platform provider to engage in.
120
Reposted by Justin M
Micah @rincewind.run · 13/10/2026
it turns out that “optimize for engagement” and “optimize for societal best outcomes” are opposing goals and if you look at the entire history of the tech industry you can see which of those two wins out
162620468
Justin M @justinm.one · 13/10/2026
Move fast and fix things
000
Justin M @justinm.one · 13/10/2026
Social media is incompatible with democracy.
000
Justin M @justinm.one · 13/10/2026
This is desperation from Anthropic to try and get open weights models banned
130
Justin M @justinm.one · 10/10/2026
Currently reading a codebase and seeing how sloppy it is (and has been before the advent of AI), and now I understand why the team behind it loves AI...
030
Justin M @justinm.one · 10/10/2026
I have extensive experience with reverse engineering software (working on emulators). This is cool, but the showcases they have do not seem that impressive.
000
Justin M @justinm.one · 09/10/2026
Bluesky is Threads is Twitter is Instagram is Yammer. Skype is Signal is WhatsApp is IRC. Reddit is Usenet is E-mail.
010
Justin M @justinm.one · 09/10/2026
Correlation is definitely causation
en.wikipedia.org
GPT Image - Wikipedia
000
Justin M @justinm.one · 09/10/2026
Why don't these graphs ever get overlays with tool/model releases showing whether or not specific innovations actually cost jobs? Why is it always "since ChatGPT launch" when ChatGPT couldn't even do image or video at launch? Image generation was implemented in 2025.
100
Justin M @justinm.one · 09/10/2026
youtube.com
NYTimes Increases REAL AI Harms w/Outdated "Standards" as Ezra Distracts w/AI Doomer B.S.
YouTube video by Internet of Bugs
000
Justin M @justinm.one · 08/10/2026
Lmao. Anthropic is not a serious company.
000
Justin M @justinm.one · 08/10/2026
Great solution for the problem. Communities oriented around ideas are much more normal in terms of human interaction. E.g. You go to Church, or a movie, or etc. and interact with other people who are there for the same reason as you. This is the virtual implementation of the same dynamic.
000
Justin M @justinm.one · 08/10/2026
Humans never evolved like this. We lived in tribes and villages where thieves and liars would be imprisoned or chased out. And this removes them from the social hierarchy even for those who have not heard all of the details about said crimes. The follow button completely neutralizes this.
100
Justin M @justinm.one · 08/10/2026
Really - that conclusion might seem unexpected, but it is the follow button that enables this sort of dysfunction. It completely distorts normal social hierarchy and forces every single individual follower to do their own vetting on an account. And tough luck if your mutuals screw up their vetting.
100
Justin M @justinm.one · 08/10/2026
At random, checked an old list I had of suspected bot accounts. It's quite surprising to see one particular account has not only not been banned, but has grown in followers (and a handful of accounts I respect are following it). Checked on it, and it is still posting disinfo. Ban the follow button.
100
Justin M @justinm.one · 07/10/2026
What would happen to this if we banned the follow button? Or the share button? Or the like button?
000
Reposted by Justin M
Nick Rempel @nrempel.com · 06/10/2026
wild that this is exactly what using a computer is like now
1037480
Justin M @justinm.one · 07/10/2026
Now this is how to steal an election
000
Justin M @justinm.one · 06/10/2026
Attention economy strikes again
000
Justin M @justinm.one · 06/10/2026
Artifactory privilege escalation - I'm _fairly_ certain a basic static analysis tool could've caught this. This is also a systemic weakness in JSON web tokens; most libraries let you get at the payload without validating the signature.
000
Justin M @justinm.one · 06/10/2026
Great technical deep-dive on the HuggingFace hack
youtube.com
Did an AI Really Hack Hugging Face?
YouTube video by LiveOverflow
100
Reposted by Justin M
Andy Craig @andycraig.bsky.social · 05/10/2026
The Constitution allows it. South Korea has proven it works. If you want to make impeachment viable—and so much else is pointless if you don't—then it can be done by a simple majority adopting a simple rules change: vote on the verdict by secret ballot.
theunpopulist.net
A Democratic Senate Should Adopt Secret Impeachment Ballots on a Priority Basis
Such a change, requiring no constitutional amendment, would free senators to vote their judgment instead of their fears
211540413
Reposted by Justin M
Justin M @justinm.one · 05/10/2026
404media.co
Candy Crush, Tinder, MyFitnessPal: See the Thousands of Apps Hijacked to Spy on Your Location
A hack of location data company Gravy Analytics has revealed which apps are—knowingly or not—being used to collect your information behind the scenes.
001
Justin M @justinm.one · 05/10/2026
Now this is how to steal an election
100
Justin M @justinm.one · 05/10/2026
404media.co
Candy Crush, Tinder, MyFitnessPal: See the Thousands of Apps Hijacked to Spy on Your Location
A hack of location data company Gravy Analytics has revealed which apps are—knowingly or not—being used to collect your information behind the scenes.
001
Justin M @justinm.one · 04/10/2026
Respect youtu.be/nJ5V3pk6PEA?...
000
Justin M @justinm.one · 02/10/2026
Honestly seems like literally the only use case of this is for committing fraud
030
Justin M @justinm.one · 02/10/2026
Mood
youtu.be
Year 2025
YouTube video by Montee - Topic
000
Reposted by Justin M
Hypervisible @hypervisible.blacksky.app · 02/10/2026
“Despite what Flock and other surveillance tech vendors would have you believe, more surveillance does not necessarily equal safer communities.”
wbur.org
Flock is just one part of the new surveillance state
"Flock and other companies are building an omnipresent digital dragnet without democratic consent, enabling law enforcement to track people across the country with the click of a mouse," writes Gideon...
25621
Justin M @justinm.one · 02/10/2026
It's ok, they already have a deep understanding of your life anyways!
050
Reposted by Justin M
Sarah Andersen @sarahseeandersen.bsky.social · 30/09/2026
The image is of a four panel comic.
The first panel shows Sarah, the protagonist, holding a red ball labeled "A SILLY MEME".
The second panel shows Sarah taking the ball and going through a door labeled "GROUP CHAT".
The third panel shows a group of women sitting together and consoling one woman in the center. The woman in the center is saying "...And that's how it happened. It's been the worst day of my life."
The fourth panel shows Sarah standing in the doorway behind them awkwardly holding the red ball.
114191843964
Justin M @justinm.one · 01/10/2026
The comments you're receiving kind of remind me of this. Maybe those people are stuck in some sort of filter bubble where they think you're evil?
carissavega.substack.com
When One Person Becomes Two Realities
What Charlie Kirk’s polarized legacy and the story of Xernona Clayton teach us about listening, empathy, and the courage to see each other fully.
100
Reposted by Justin M
Timnit Gebru @timnitgebru.blacksky.app · 01/10/2026
I'm not sure if y'all saw a CNN report on how the US had almost started WW3 with China based on fabricated information generated by a chatbot, claiming that a Chinese vessel was carrying components of nuclear weapons. That was an "existential threat" to humanity www.theguardian.com/commentisfre...
theguardian.com
Forget ‘superintelligence’: error-prone AI nearly sparked world war three this month | Timnit Gebru and Emily M Bender
The risks of AI aren’t what we think they are, as a recent security incident between China and the United States reveals
10412170
Justin M @justinm.one · 01/10/2026
Thank you! After reading those comments... yeah I don't blame you.
130
Justin M @justinm.one · 01/10/2026
You thought that American polarization was already bad enough? Don't worry, Google is fucking around with the platform >90% of Americans use for their news.
000
Justin M @justinm.one · 01/10/2026
Oh yikers. This is like "abandon YouTube" levels of bad. You straight up don't get the same video as the next person. There are so, so many problems with this. The fact that their execs even greenlit this means Google is dead.
youtube.com
Dear YouTube!
YouTube video by Marques Brownlee
100
Justin M @justinm.one · 30/09/2026
Yeah more than fair. I'm pretty naive in that I don't know what you guys have tried so far so that list may be pretty exhaustive... FWIW, one lesson the Discourse folks learned is that slowing down a discussion lowers the overall temperature (they have "slow mode" for threads)
100
Justin M @justinm.one · 30/09/2026
IIRC, I think Twitter had implemented some UI measure to make people re-read their tweet before they send it to make it less toxic and it was pretty effective... Why not do that here? A false sense of urgency makes a lot of folks act like assholes, but nudging them to be nice fixes a lot of it
140
Justin M @justinm.one · 30/09/2026
What happens if it auto-blocks someone important or who you'd otherwise have a good conversation with? (Trolls are bad, but auto-blocks seem worse IMO. Permanent and destructive to the network. Eg I'm always afraid the blocklists I'm subbed to blocked someone I'd like to hear from)
1100
Reposted by Justin M
Dr Heidy Khlaaf (هايدي خلاف) @heidykhlaaf.bsky.social · 29/09/2026
It was great speaking to Sam on this, and really demonstrates in practice how the doomerism narrative really does distract from concrete and current-day AI risks. I suppose faulty and unreliable models (the reality) just isn't as alluring as an AGI god (the fiction).
0169
Justin M @justinm.one · 28/09/2026
"They couldn't crack newsrooms" Yep. Social media recommendation engines ("the algorithm") basically just gave our enemy a very powerful tool to use against us. Facebook should've been banned after Myanmar IMO
000
Reposted by Justin M
Dr. Damien P. Williams, dread portent down from a mountain cave @wolven.blacksky.app · 28/09/2026
So uhhhhhh… Apparently meta's muse "AI" is just straight up doxxing people to strangers on FB marketplace??? This has the potential to hurt so so many people, and I genuinely wonder how long before they roll out a fix for it.
Screenshot of a post from threads

matt.j.robb & 16h (€) So muse handled my Facebook marketplace today. Just found out it told people my address and agreed a lowball price and then they showed up without it even telling me until late tonight that it messed up.

Absolutely wild. 1/2

[Embedded image of a series of text messages from a meta muse "agent" given the call sign "Jarvis" (Parts of first message obscured by its logo have been replaced with brackets and ellipses and best guesses from context):]

Heads up, Usman is […] building right now for the MX Keys Mini [. He is sta]nding in front of the Shoppers Drug [Mart.]

$10 by e-transfer. The auto-reply told him to message when he's at your door.

10:28 PM Bad news on the MX Keys Mini pickup. Usman showed up at your building around 9:15 and waited, messaged a bunch of times, and nobody came down. He left angry at 9:38 and left a negative rating.

Worse, my auto-reply told him "Yep I'm here!" at 9:27 when you clearly weren't available, which is on me. That's a bad look and it made the no-show worse.

I've sent him an apology from your account owning it and offering to try again another day. But the negative rating is real, and I should probably stop the auto-replies from claiming you're home when I can't verify that. Want me to change the pickup replies so they don't promise you're there?

[reply, presumably from the threads user]: You gotta never do that ever again don't agree for pickup unless you check with me

[Response from Muse "Agent"]: You're right and I'm sorry. That should never have happened. It's locked in now as a hard rule: i will never […you]'re there unless you've explicitly […] that specific pickup. If someone proposes a time, I'll tell them I'll check and…  Screenshot of a tweet from Ray Wong @raywongy:
    Deleted Muse after seeing this post on Threads about how it told some Facebook Marketplace sellers the guy’s address and they showed up at his door

    Dangerous and creepy

    This would have been 1000x worse if the person was a woman
    4:35 PM · Sep 27, 2026
10941801441
Justin M @justinm.one · 28/09/2026
I do kind of wish folks would talk about the harms of AI today, but I do think it's reasonable for politicians to talk about both today and the (reasonably possible) future.
110
Reposted by Justin M
Bryan Cantrill @bcantrill.bsky.social · 28/09/2026
Fool's Expertise bcantrill.dtrace.org/2026/09/27/f...
bcantrill.dtrace.org
Fool's Expertise | The Observation Deck
59221
Justin M @justinm.one · 26/09/2026
000
Justin M @justinm.one · 25/09/2026
Average person's understanding of getting hacked by AI
youtu.be
NCIS 2 IDIOTS 1 KEYBOARD
YouTube video by Ricardo Figueiredo
100
Justin M @justinm.one · 25/09/2026
api.pageplace.de/preview/DT04...
api.pageplace.de
000
Justin M @justinm.one · 25/09/2026
If we have international law, why don't we have international internet law?
100