Taggart :ifin:

Displaced Philly boy. Threat hunter. Educator. :ifin: Executive Director. #infosec, #programming #rust :rust:, #python :python: #haskell :haskell:, and #javascript :javascript:. #opensource advocate. General in the AI Resistance. Runs @thetaggartinstitute. Made https://wtfbins.wtf. Not your bro. All opinions my own. Dad. #fedi22 #searchable

Pronouns: He/him.

Public Key

npub18wjp9tztznztxlxka5ttn5nz448la7c9ckmvdvlptcupgud3ygdqj6q6gg

NIP-05 Address

mttaggart@infosec-exchange.mostr.pub

Profile Code

nprofile1qqsrhfqj4393f39n0ntw694e6f326nl7lvzutdkxk0s4uwq5wxcjyxspz4mhxue69uhhyetvv9ujuerfw36x7tnsw43qzn4y35

Publishing to

relay.ditto.pub

Author Public Key

npub18wjp9tztznztxlxka5ttn5nz448la7c9ckmvdvlptcupgud3ygdqj6q6gg

Show more details

Published at

2026-07-17T02:17:10Z

Kind type

0 Metadata

Event JSON

{ "id": "0d7a89b0cb434d0a8b82f2cdef8462f8000f7fa72b8ef9e4b9f5706ac554815f", "pubkey": "3ba412ac4b14c4b37cd6ed16b9d262ad4ffefb05c5b6c6b3e15e381471b1221a", "created_at": 1784254630, "kind": 0, "tags": [ [ "emoji", "ifin", "https://media.infosec.exchange/infosec.exchange/custom_emojis/images/000/663/935/original/d4a896a001495b7f.png" ], [ "emoji", "rust", "https://media.infosec.exchange/infosec.exchange/custom_emojis/images/000/029/155/original/c0a6c76c4e012ab5.png" ], [ "emoji", "python", "https://media.infosec.exchange/infosec.exchange/custom_emojis/images/000/029/197/original/3a384cedbffde920.png" ], [ "emoji", "haskell", "https://media.infosec.exchange/infosec.exchange/custom_emojis/images/000/028/818/original/35ece0354386bf51.png" ], [ "emoji", "javascript", "https://media.infosec.exchange/infosec.exchange/custom_emojis/images/000/029/136/original/02529a2e0c7ee0f6.png" ], [ "proxy", "https://infosec.exchange/users/mttaggart", "activitypub" ], [ "client", "Mostr", "31990:6be38f8c63df7dbf84db7ec4a6e6fbbd8d19dca3b980efad18585c46f04b26f9:mostr", "wss://relay.ditto.pub" ] ], "content": "{\"name\":\"Taggart :ifin:\",\"about\":\"Displaced Philly boy. Threat hunter. Educator. :ifin: Executive Director. #infosec, #programming #rust :rust:, #python :python: #haskell :haskell:, and #javascript :javascript:. #opensource advocate. General in the AI Resistance. Runs @thetaggartinstitute. Made https://wtfbins.wtf. Not your bro. All opinions my own. Dad. #fedi22 #searchable\\n\\nPronouns: He/him.\",\"picture\":\"https://media.infosec.exchange/infosec.exchange/accounts/avatars/108/196/880/698/187/602/original/0cef8947c107a7a8.jpg\",\"banner\":\"https://media.infosec.exchange/infosec.exchange/accounts/headers/108/196/880/698/187/602/original/1a834b355f8a1f41.jpg\",\"nip05\":\"mttaggart@infosec-exchange.mostr.pub\",\"fields\":[[\"The Taggart Institute\",\"https://taggartinstitute.org\"],[\"Blog\",\"https://taggart-tech.com\"],[\"Codeberg\",\"https://codeberg.org/mttaggart\"],[\"YouTube\",\"https://youtube.com/taggarttech\"],[\"GitHub\",\"https://github.com/mttaggart\"],[\"Keyoxide\",\"aspe:keyoxide.org:G4ADJFWICZZZXGR4STZQVMBJNM\"]]}", "sig": "b94543a34ea8c8887fe3a4a55aaa1712c83d469684e182133219ddbfdb82e1b2e48bce910f713f5d58154140371fea688152dbfe70f0bea21db7e481fe9b9579" }

Last Notes

2026-05-26 18:38:34 UTC

- reply

@nprofile…fqfm There will be someone in this thread that comments how, if this is the experience, we're "doing SSO wrong." Okay but I've never seen it done right.

2026-05-26 18:35:38 UTC

I swear I lose about a day per month to terrible auth flows that derail my thought process and send me someplace besides what I was trying to log into in the first place.
I know my brain is medium to extra spicy, but I can't be the only one with this struggle.

2026-05-23 19:46:28 UTC

- reply

It can be simultaneously true that intellectual property law/IP law is ridiculous and that stealing people's stuff is wrong! Those are not in conflict, and I don't really think "Well it's legal" is that rad of a defense.

2026-05-23 19:40:39 UTC

I'm not sure why, but there's something about intellectual property that makes folks struggle with legality versus morality. I suspect reasons range from the desire to do piracy without reprisal to ardent belief that there is no such thing as "intellectual property."

2026-05-20 17:00:12 UTC

Antiques Roadshow but it's all old computers and game consoles.

2026-05-16 19:54:56 UTC

This entire report from the Ontario government on genAI systems is worth a read, but the review of healthcare scribe accuracy is pretty devastating, imo. This has to work for the tech to be worth anything. If the notes in the chart are wrong, the whole thing falls apart.
https://www.auditor.on.ca/en/content/specialreports/specialreports/en26/2026_AI_EN.pdf
https://media.infosec.exchange/infosec.exchange/media_attachments/files/116/586/007/548/909/881/original/43a301bc8c0b6234.png

2026-05-12 18:39:32 UTC

- reply

@nprofile…0pvp We have multiple reports of folks receiving Canvas-themed phishing lures in the last day or so.

2026-05-12 18:36:11 UTC

Please get in touch. Something ain't right here.

2026-05-10 23:42:24 UTC

- reply

@nprofile…5a85 @nprofile…65qj Show me a stat for 90% and I'll show you a made-up stat.
The code generation "works," as in, is syntactically correct and can compile. It better! Code is a highly structured, oversampled language in the training corpora. If it can't generate that, it's hopeless.
But would you trust a bridge designed by a LLM? How about a medical treatment plan? Or a business contract?
The fallacy tech keeps making is that any competency with code implies broader application.

2026-05-10 22:04:46 UTC

The constant mental vigilance in a generative world is exhausting.
"I asked Claude to do $thing and it did this!"
No it didn't. No you didn't. Probably none of that happened.
And somehow, being unwilling to admit the thing is just making stuff up is annoying and unnecessary, not the damn model.

2026-05-07 17:18:40 UTC

- reply

@nprofile…c808 Wild. For the JSFile one it is SYSTEM. But okay.
So HCKR\.URL\OpenWithProgIds. Delete InternetShortcut and add anything, including JSFile, right?
https://media.infosec.exchange/infosec.exchange/media_attachments/files/116/534/432/193/996/170/original/892314441db1f9a7.png

2026-05-07 17:05:52 UTC

- reply

@nprofile…c808 I mean doing stuff as SYSTEM is not impossible. And as far was what it targets, that's fairly discoverable. .URL references InternetShortcut

2026-05-07 16:57:14 UTC

- reply

@nprofile…c808 Wait why doesn't going after HKCR\InternetShortcut\Shell\Open\Command do it?

2026-05-07 15:04:02 UTC

- reply

@nprofile…c808 We moved the handler of .js files from Windows Script Host to Notepad to neuter JS-based malware. Would a similar mitigation make sense here? I'm not aware of any real-world usage of .url files.

2026-05-07 01:40:51 UTC

Some optimism as a counterweight to, well, everything.
https://terrygodier.com/the-boring-internet

2026-05-06 17:41:22 UTC

- reply

@nprofile…r4z3 I do not.

2026-05-06 16:43:38 UTC

- reply

I failed 8th grade algebra. Straight up F. I was told I was bad at math through high school and college, and believed it. I barely got through remedial precalc in my senior year. As an English major, I fulfilled my "quantitative reasoning" course requirement by taking Formal Logic, which I aced.
Then I taught myself programming. Then I taught myself calculus, partially through programming. It was then I realized that I suck at arithmetic, but not math. In fact, I had barely been taught math but for arithmetic.

2026-05-06 16:40:23 UTC

- reply

If you think it's ridiculous that writing is judged on spelling instead of content, you also think manual arithmetic is a poor measure of mathematical understanding. They are distinct skills, and have been conflated for far too long.

2026-05-06 16:38:56 UTC

One of my most fervent education takes is that we should be teaching programming earlier, and more than manual arithmetic. By high school, no student should be carrying the 1 on a math test.

2026-05-04 20:13:40 UTC

- reply

@nprofile…r3m4 I know I have tech journalists following me. I don't know why this isn't a bigger story. A part of the internet considered "Critical infrastructure" is being attacked with impunity, and those who could stop it are doing nothing.

2026-05-04 20:11:39 UTC

- reply

@nprofile…r3m4 10/10 excellent brand synergy
https://media.infosec.exchange/infosec.exchange/media_attachments/files/116/518/125/437/313/040/original/abfb3d26b06b6e09.png

2026-05-04 19:45:29 UTC

Hey @nprofile…r3m4, how come you're protecting beamed[.]st, the DDoS service that's attacking Ubuntu? It's an obvious criminal enterprise that literally advertises botnet access.
https://media.infosec.exchange/infosec.exchange/media_attachments/files/116/518/022/518/570/386/original/f50b1d79e2610926.png

2026-05-04 18:04:54 UTC

This incident is going to turn me into the Joker. Canonical has said nothing since Friday.

2026-05-04 13:47:10 UTC

- reply

@nprofile…f0us Ironic to hear you say that! Interestingly, I don't think Debian is the move in a world where we expect more 0-days and N-days at rapider paces of exploitation. I've been very happy with Fedora Server, actually.

2026-05-04 13:41:26 UTC

I promised myself I would leave this alone but it's Monday morning and after finally trying to update my Ubuntu servers, I'm getting nothing but timeouts.
Something is still not right over there, and I'm pretty annoyed. Annoyed enough to consider not using Ubuntu as a server base moving forward.

2026-05-03 14:19:41 UTC

I feel like I'm taking crazy pills. Three days into service interruptions, with a major patch to install, and no word from Canonical. It's a shocking abdication of responsibility of care for their users.
If anyone who knows wants to share with me confidentially what's going on, DMs are open.

2026-04-30 22:08:26 UTC

Got a very silly #CopyFail container escape working.
Basically, if the container can see a file shared by the host, regardless of permissions, CopyFail can write to it on the host.
https://discourse.ifin.network/t/copy-fail-732-bytes-to-root-on-every-major-linux-distributions/342/44

2026-04-29 18:49:57 UTC

Yeah I tested CopyFail. It's real. Yikes.
https://discourse.ifin.network/t/copy-fail-732-bytes-to-root-on-every-major-linux-distributions/342

2026-04-29 16:01:46 UTC

Ugggh this one is going to suck. cPanel is everywhere and most are not patched frequently.
https://www.bleepingcomputer.com/news/security/cpanel-whm-emergency-update-fixes-critical-auth-bypass-bug

2026-04-28 13:05:15 UTC

May I suggest...not exposing your robot control plane to the internet
https://www.resecurity.com/blog/article/cve-2026-25874-hugging-face-lerobot-unauthenticated-rce-via-pickle-deserialization

2026-04-27 23:03:39 UTC

- reply

Coming at this another way: MIT cites 50 GWh for training GPT-4. Depending on US zip code, that's going to be somewhere between 11-20 metric tons of CO2 equivalence. That's 1000x less than the cited Mistral number.
So we have to conclude that training is bad, but given current usage rates, inference will blow away that number in short order, so don't let anyone cite per-inference costs as a stat in LLMs' favor.
https://www.technologyreview.com/2025/05/20/1116327/ai-energy-usage-climate-footprint-big-tech/

2026-04-27 22:33:04 UTC

- reply

Seems like they have conflated training and 18 months of inference. So the question is how much of this number was pure inference, and what was training.

2026-04-27 20:43:41 UTC

- reply

Okay so if I did this right with the EPA's converter, training/using (marginal) Mistral was
https://media.infosec.exchange/infosec.exchange/media_attachments/files/116/478/615/215/201/384/original/f1e04d3163f41fc8.png

2026-04-27 20:37:59 UTC

- reply

Oh yeah! Mistral did publish some data, for what it's worth:
https://mistral.ai/news/our-contribution-to-a-global-environmental-standard-for-ai

2026-04-27 20:15:19 UTC

Has any "open" model published data on the energy requirements for training? It sure would be nice to know those sorts of details.

2026-04-25 21:58:29 UTC

A URL mode for mobile keyboards. Now that I've named it, I want it so badly.

2026-04-25 13:41:25 UTC

- reply

Related reporting here: https://www.404media.co/the-ai-compute-crunch-is-here-and-its-affecting-the-entire-economy/

2026-04-25 13:27:56 UTC

Buried inside this excellent article is a fact that might surprise some:
“If you have thousands or millions of people using [coding models] every single day, the inference costs of just the users generating tons and tons of tokens all the time really outweighs the training side of things.”
The reality of reasoning models is that ongoing inference is even more energy-intensive than the training itself.
But also, yeah, nobody knows how this is getting paid for without a gigantic price hike, and the cost/benefit for a lot of businesses is about to become obviously toxic.
https://www.theverge.com/ai-artificial-intelligence/917380/ai-monetization-anthropic-openai-token-economics-revenue

2026-04-22 05:39:05 UTC

- reply

Lmaooo
Using Chronicle increases risk to prompt injection attacks from screen content. For instance, if you browse a site with malicious agent instructions, Codex may follow those instructions.
One wonders how many things it will OCR into a prompt injection. Utter clownshow of a technology.

2026-04-22 01:51:42 UTC

- reply

@nprofile…qqrn If you're objection is extractive mining or the oppressive regime of China as your barrier to purchase, those are reasonable objections.
You can find no computing device, including Framework, without them. Whence the rare earths?
So yeah, it all sucks. What I want is the company to own up to the mistake. I'm putting the onus on the company. People can make the purchasing decisions that make sense for them.

2026-04-22 01:47:31 UTC

Oh my god, OpenAI reinvented Recall, but for macOS.
"Chronicle" is an opt-in feature that scans your screen, saves screenshots temporarily, and sent to OpenAI's servers.
https://developers.openai.com/codex/memories/chronicle

2026-04-22 01:26:46 UTC

- reply

Lotta "Well he's not actually a Nazi replies. Look:
The term is loaded. It absolutely is. But if, in this age, you have vociferously advocated for
checks notes
racial purity in your public writing, I think you surrender the benefit of the doubt. And I'm not waiting for you to line people up before I decide it's a problem.
More simply: racist scum don't need your advocacy.

2026-04-21 22:23:49 UTC

- reply

@nprofile…qqrn We don't have to accept one at the cost of the other! There's no structural reason why Framework has to support crappy people while pursuing their mission. They decided to do so, and people can choose how to reduce e-waste in a lot of other ways besides buying one laptop brand.

2026-04-21 19:31:21 UTC

- reply

Listen, I have a Framework. It's a great laptop—maybe the best I've ever had. But I can't support a company that funds, in any form, an actual real-life Nazi like DHH. I also need them to, y'know, take ownership of the massive misstep doing so is/was.

2026-04-21 19:20:15 UTC

Yo did we forget about the Framework-funding-fash thing or

2026-04-21 17:28:13 UTC

Okay it's funny, but also employees doing stuff like this on their work machines is the easiest way to get popped.

2026-04-19 03:28:10 UTC

- reply

@nprofile…kzr2 I mean LLM generated code, and no, they are not the same, but the argument about a "skill issue" is.

2026-04-19 02:55:39 UTC

I wonder what the overlap is between developers who think there's nothing wrong with C/C++ because "if you're good" you can write memory-safe code at scale, and developers who believe you can make generative code safe and secure with the "right guardrails."

2026-04-17 22:39:19 UTC

More coherent than IPv8
https://www.ietf.org/archive/id/draft-meow-mrrp-00.html

2026-04-16 19:48:22 UTC

Very amused that my lil Proxmox cluster has better uptime than Bluesky.

2026-04-16 18:29:27 UTC

- reply

@nprofile…k987 Nope, it's not structured like that. This targets OS creators directly, and ties violations to FTC rules about fraudulent claims.

2026-04-16 18:08:40 UTC

I've read the (early) text of HR 8250, the "Parents Decide Act." It is nearly structurally identical to California's law. Actually, it is less rigorous, as the legal obligation is entirely on OS creators, not app developers.
Again, now is the time to contact reps to oppose this legislation. The term definitions are wacky and unworkable, for one thing. But also, it just is silly and unenforceable in general. Watch for amendments, but don't accept this as a done deal.
https://www.congress.gov/bill/119th-congress/house-bill/8250/text

2026-04-16 13:31:11 UTC

- reply

@nprofile…q2xw I'm super skeptical on both counts. First, the entire purpose of the functionality here is to provide summarization. That can't be deterministic because ti's the core function of the LLM itself. Cases have been overturned because law enforcement relied on AI summaries that hallucinated.
As for the second point, my contention is not that the evidence won't be admissible. I agree that courts will accept it as evidentiary. My contention is that opposing counsel will go hog wild with it as a way to produce reasonable doubt.

2026-04-16 12:01:38 UTC

- reply

Uhhh also this tool is designed for proprietary, corpo, cloud-hosted models. Dunno about y'all but my forensic evidence can't leave the lab in any form without chain-of-custody procedures. What are we doing here?

2026-04-16 11:55:17 UTC

Imagine testifying in court that you relied on a forensic tool known to make stuff up.
Defense attorneys are gonna love this one.
https://www.autopsy.com/autopsy-4-23-0-release-claude-ai-assistant-mcp-cyber-triage-integration/

2026-04-16 01:08:21 UTC

I love these fixes because it means that: a) I get to learn something new about the OS, and b) we get to confound the adversary.

2026-04-15 15:07:43 UTC

- reply

I really think we have a fundamental disconnect about motivations on this legislation. This community is rightfully concerned about issues like a surveillance state, trans genocide, authoritarianism, etc. I will humbly submit that is not the motivation of these bills. Instead, the general public loathes big tech—and social media in particular. Note the verdicts coming down against Meta. The (correct) narrative is: Big Tech made products that harmed kids. The voting public wants "something" done about it. It's a bipartisan issue, an obvious legislative W for lawmakers who struggle to get anything through either chamber.
IMO, that's what is driving this, not the aforementioned evil designs. Those are still there, but I don't think this is a direct manifestation of them.

2026-04-15 13:45:05 UTC

I wrote a bit about the CA law, and I am of the opinion that it's silly and unenforceable as written, but not the end of the world. I also think it was important that the Governor (who sucks) basically begged for amendment before it went into effect. I suggest an exemption for open source. The text of the Congressional Bill has yet to be introduced, but if your reps are on the Energy and Commerce Committee, now's the time to make noise to them.
https://taggart-tech.com/ab-1043/
https://energycommerce.house.gov/representatives

2026-04-15 13:36:14 UTC

- reply

@nprofile…7yuj FWIW, the California legislation was opposed by big tech's lobbying firms. IMO what we have here is a fundamental disconnect between technical people and non-technical people in understanding how to address a perceived issue (the internet is scary for kids). Polling consistently shows broad support for this issue, which is in part why this is a bipartisan bill.
I'm not agreeing with it; I'm just saying the motivations might not be from Big Tech itself.

2026-04-15 13:31:51 UTC

- reply

@nprofile…rh4q "What if workflow automation...but sometimes it can go off the rails and there's no way to know when that will happen?"
"Ship it!"

2026-04-15 00:41:37 UTC

- reply

@nprofile…ytky @nprofile…ckj7 The second stage did indeed have a handful of crypto exchanges it would try to hit. No reason not to block them all in an enterprise environment, imo.

2026-04-14 19:18:30 UTC

I'm immune to LLM sycophancy because I'm terrible at taking a compliment.

2026-04-13 18:23:18 UTC

- reply

@nprofile…n6tl Ended up giving basically a TED talk at a toddler's birthday party because The Dads (non-tech people) were convinced it was AGI.

2026-04-13 16:22:02 UTC

New York Times, I beg you to crack open one (1) book that concerns dates prior to 1900.
https://media.infosec.exchange/infosec.exchange/media_attachments/files/116/398/314/147/324/332/original/b1ca2848d869214e.png

2026-04-13 13:12:39 UTC

This cost, which somehow nobody wants to talk about, is an erosion of the foundation of our society. Faster code generation is simply not worth this cost. What's worse, it's self-defeating: A generation educated by generative models will be ill equipped to build new things.
https://arstechnica.com/science/2026/04/to-teach-in-the-time-of-chatgpt-is-to-know-pain/

2026-04-12 21:37:51 UTC

It was my job for a decade to try to keep tweens and teens safe online. Let me tell you what you already know: no law or technology can do it. There are no lengths kids won't go to to talk to friends without prying eyes. The harder you try to lock it down, the dodgier their solutions will be.

2026-04-11 22:10:54 UTC

If you're forced to use gen AI at work, there's a really useful government manual from the 1940s that might might be useful. I'm sure you can find it.

2026-04-10 15:48:31 UTC

I don't usually link Zitron, but this one is important. It isn't the tech that will derail this hypetrain; it's the money.
Nobody can actually afford what's going on right now. Just like in 2008, it's sleight-of-hand that will eventually be revealed, and that's when gravity takes over.
If you are building anything that assumes these services will be around and/or price-stable, you are recklessly adding liability to your organization.
Tech has always had a problem with viable business models, and this is the final form of that monster.
https://www.wheresyoured.at/the-subprime-ai-crisis-is-here/

2026-04-10 13:55:30 UTC

- reply

@nprofile…rues Why are TPMs are deal breakers on Linux?

2026-04-10 13:35:02 UTC

- reply

@nprofile…vs6a I can already see it happening, and it's already driving me crazy. For people so interested in math, counting (qubits) seems remarkably challenging for them!

2026-04-10 13:23:07 UTC

When AI pops, I'm looking forward to watching LinkedIn eat itself alive searching for a take.

2026-04-10 06:17:40 UTC

A demo/production Ringspace server is finally live!
https://ringspace.taggartinstitute.org
I'm actively looking for infosec blogs who'd like to join. I have a feeling I know some folks...

2026-04-09 17:26:18 UTC

Near as I can tell, this is all very good news. More things should take advantage of secure enclaves, and this open standard protects against one of the hardest current defense surfaces.
https://security.googleblog.com/2026/04/protecting-cookies-with-device-bound.html

2026-04-09 01:07:43 UTC

- reply

@nprofile…6g65 Agreed, but I'm not sure you can have one without the other. At least not until these giant trainers get swallowed by the earth.

2026-04-09 00:34:08 UTC

Yeah so let's refocus on the real perspective here. It doesn't matter at all how good these models are at code or finding vulnerabilities if we destroy our ability to seek and share knowledge.
https://futurism.com/artificial-intelligence/google-ai-overviews-misinformation

2026-04-08 16:40:54 UTC

- reply

@nprofile…ktd9 Looks at the rare earths in...every device
Uh

2026-04-08 16:35:35 UTC

- reply

But again, don't imagine for a second gen AI tools are the only ones with externalities. Be honest with yourself about what else you're tacitly accepting.
There is no perfect here, only trying to do better.

2026-04-08 16:22:24 UTC

- reply

@nprofile…qml3 No but I am crabby 🦀

2026-04-08 16:16:13 UTC

Hey for what it's worth, I've been writing my own Rust for the past three days getting Ringspace ready for production use. I don't want a sticker or anything, but, y'know.
I can still do it.

2026-04-08 15:56:58 UTC

I agree with this framing wholeheartedly. This is a succinct and accurate description of the assumed externalities when you choose these tools.

2026-04-07 17:55:16 UTC

Aww yiss, @nprofile…ckj7 logos available here on infosec.exchange! Thanks @nprofile…qml3 !
:ifin: :ifin_i:

2026-04-07 16:30:50 UTC

- reply

@nprofile…t9a9 Might be something blocking. I have gotten it to go away, and we did test it with folks prior to launch.

2026-04-07 15:32:03 UTC

This has been a year in the making. As Executive Director of IFIN, I'm very excited for what's coming. Join us!

2026-04-07 13:19:32 UTC

- reply

@nprofile…vs6a I'm SAYIN

2026-04-07 13:17:10 UTC

This is the one.
Although I would have preferred it not be called "Earthset," good grief.
https://images.nasa.gov/details/art002e009288
https://media.infosec.exchange/infosec.exchange/media_attachments/files/116/363/613/331/709/915/original/d18dadb8744c41f4.png

2026-04-06 20:35:59 UTC

When they update the diceroller, diceroll-reliant processes go sideways.
https://github.com/anthropics/claude-code/issues/42796

2026-04-06 18:17:04 UTC

- reply

Bluesky is very busted, not sure what else.

2026-04-06 17:14:52 UTC

Looking like an incipient us-east-1 issue

2026-04-06 17:06:58 UTC

- reply

@nprofile…e09a This toot brought to you by MBAs, but same thing

2026-04-06 16:58:39 UTC

- reply

I've been burned so many times; I've learned my lesson. You really need to read each of these things carefully if you want to understand what the researchers are concluding. Reading a news article—even worse, just the headline—is at best no information, at worse disinformation.

2026-04-06 16:57:39 UTC

I am begging AI researchers trying to study human impact to get very rapidly better at methodology so I don't constantly read halfway through these papers only to find some ridiculous experiment design that will throw the conclusions into the air.

2026-04-06 15:31:15 UTC

Best mouse ever

2026-04-06 15:27:26 UTC

- reply

@nprofile…vs6a No no, not @nprofile…c2pe acquiring @nprofile…scke. The other secret project.

2026-04-06 15:22:46 UTC

- reply

@nprofile…rh4q On purpose

2026-04-06 15:11:42 UTC

Big secret project a year in the making launches tomorrow.

2026-04-06 07:20:29 UTC

Didn't really think I needed to say this, but equating gen AI and its uses to the human horrors of slavery is actually deranged. If you have a take like that, keep it to yourself and go sit in the corner, maybe forever.

2026-04-05 01:56:23 UTC

- reply

@nprofile…uqdf @nprofile…7yuj I think that's very narrowly right for creative preferences, ignoring the much larger reasons folks don't like them. You can be an outcome person and still find them dangerous, unethical, etc.

2026-04-03 20:15:49 UTC

Another Achewood strip becomes reality.
https://achewood.com/2006/05/22/title.html
https://media.infosec.exchange/infosec.exchange/media_attachments/files/116/342/610/161/167/970/original/3204f1acc6c27c38.png

2026-04-03 17:57:12 UTC

- reply

@nprofile…02yd Hey @nprofile…7yuj ☝️ this was really bothering me. If the current precedent stands, it's absolutely the case that no open source license is enforceable on generative code, as the copyright is a prerequisite for any license.
I imagine there's a test of amount still, like if most of the code is human-authored, you could still claim copyright. But for example, the tool I just made with Claude Code as an experiment? Full public domain, no terms available to me.

2026-04-03 17:46:04 UTC

- reply

@nprofile…02yd
Can you license something without owning the copyright?
No, you need to own the copyright or have permission from the copyright holder to license the work.
So if no copyright is possible, no license is possible?