Adam Lasnik

Adam Lasnik

@adamlasnik.net

Reasonably social geek based in the SF Bay Area. Previously worked to improve healthcare access while a Program Manager at Google. Now leading operations at a music startup. www.AdamLasnik.net

772 followers517 following791 posts

Wow... always nice to see some good news that counterbalances the worrisome trends in France, Canada (BC), and other places!

The long-serving strongman leader of Serbia, Aleksandar Vucic, has conceded defeat in an election that pitted him and his right-wing party against a diverse student-led opposition movement that spent two years protesting his government.

Has journalism indeed gotten way worse the last few years, at least headlines-wise? e.g., "Bay Area Restaurant Investigated for Major Issue" when the responsible headline should read: "[Restaurant name] in [City] Investigated for [Actual issue]" I expect it's purely to incent clicks, but...

...I'm gonna either - not click OR - click, see it's for a very far away city, and instantly nope out OR - get so annoyed with this outlet that I never again click on their articles Maybe their metrics show this shittiness is -- at least in the short term -- profitable? But just seems wrong

This statement, sadly, is pretty evergreen in social media.

Zach Weinersmith@zachweinersmith.bsky.social

Mostly I'm laughing about this but there really is something sick at the heart of our epistemology on this site sometimes. Probably it was always like this, but things aren't true because your ideology demands it. All you have to do is read.

"But...but... what if Trump tampers with the elections? What if he doesn't honor them?" These are possibilities, sure. But the MOST POWERFUL thing we could do NOW is to forkin' vote and encourage others to vote as well. It's much easier to manipulate, litigate, or deny close vs. blowout races!

Personal note: Sometimes little things really help :) When waiting for water to boil or cooking stuff in my microwave, I've been working to improve my balance & stability by standing on one foot at a time and positioning my leg back and forth. It's made a surprising dif in weeks! Highly recommend!

Adam Lasnik reposted

Replying to @masnick.com

One of the most humbling facts from my time in Community Ops at Meta was the realization that, at 99% accuracy, we made more than a million moderation mistakes a day. Purely because of the scale of the platform. You can’t six sigma the work because the cost would be unfathomable for any company.

Adam Lasnik reposted

Replying to @aaronhm.bsky.social

I feel like I'm taking crazy pills, but am I the only one who feels it's EASIER than ever to find awesomely useful and interesting stuff on the internet? Yes, there's orders of magnitude more crap, but there are ALSO tuvan throat singing vids and info on how to fix any home appliance...

The majority of replies in this thread are disheartening exhibits of conflating WISHED reality vs ACTUAL reality. Also, arguing with Mike about sec 230 and the First Amendment is like engaging in a land war in Asia. Bad idea! (Because he really knows his shit on this!)

Mike Masnick@masnick.com

People demanding Zuck be hauled before Congress to explain why No Kings insta was suspended... how would you feel if MAGA GOP hauled MSNow execs to explain why they covered Trump the way they do? Both are editorial choices protected by the First Amendment. If this annoys you...

Related: it’d be awesome if we as a community were as emphatic about personally encouraging individuals to be a part of Bluesky as we are virtue signaling our departure from Meta properties. (“Yeah? Well *I* quit Instagram in 1993! What took you so long?!”)

Adam Lasnik reposted

Kind of stunned how many people in my mentions still believe that the point of Section 230 is that if you edit, you lose its protections. That's literally the exact opposite of why the law was written and the exact opposite of what courts have said. But tons of people here believe it. Weird.

techdirt.comHello! You've Been Referred Here Because You're Wrong About Section 230 Of The Communications Decency ActHello! Someone has referred you to this post because you’ve said something quite wrong about Section 230 of the Communications Decency Act. I apologize if it feels a bit cold and rude to resp…

MacOS 27 has been out for more than a month, and the vast majority of audio plugins & tools are STILL not vetted as compatible (!) 😡 www.production-expert.com/apple-macos-... This seems different than other sw swgments. Is that right, or is the music tools space just really behind, & if so, why?

production-expert.commacOS Golden Gate Audio Compatibility: DAWs, Plugins and HardwareIs your DAW, plugin or audio interface compatible with macOS Golden Gate? This regularly updated chart covers Pro Tools, Logic, Studio One, and key audio hardware.

Adam Lasnik reposted

Everyone I don’t like is a bot/Russian agent/conspiratorial bad actor is a pretty useless way to sort through online interactions

Tyler, from the Internet (Haunted)@tylerfromtheinternet.com

Matt Darling, Matthew Yglesias, Sky Mariachi/Will Stancil on here, Economist writers, others like them: I think it's MORE embarrassing if they AREN'T actually being paid by Russia to be such insufferable poison-pill dipshits

Adam Lasnik reposted

Replying to @maxdubler.com

I saw someone the other day remark that San Francisco is the only city in the country where the left is reactionary and the center is progressive.

Laws/lawsuits "...for the children!!" nowadays are depressingly so often performative at best, harmful to children at worst.

I generally avoid podcasts. Life's too busy to not... just-read stuff. Except I got sucked into this one. It's about math and AI, sure, but really focuses on purpose and meaning and includes mentions of meth, sex, God and fries. www.nytimes.com/2026/10/13/p... It's genuinely a MUST-LISTEN.

nytimes.comWhen A.I. Comes for Your Purpose in LifeIn the span of just a few months, A.I. systems have solved some of the biggest outstanding problems in math — adding mathematicians to the list of those terrified of losing their job to the technology...

But if you are, like I've been, adamantly opposed to podcasts, well, I've got you covered. Here's a 'Clif Notes' version, complete with awesome pull quotes. Made, of course, by AI. claude.ai/artifact/PgU... And here's the whole darn transcript daily-math-human-purpose.thatadamguy.chatgpt.site

claude.aiBeautiful in Its TimeAn illustrated long-read of The Daily's Oct 13, 2026 episode with Steven Strogatz on AI solving math's hardest problems, with pull quotes and a live cricket-synchrony simulation.

Still, I'm serious, though. Go listen to this one. It made the host tear up. And the math prof. And me. And will probably get to you, too. It's depressing AND hopeful. www.nytimes.com/2026/10/13/p... If you don't believe me, skim the comments, a great many of which are equally enthusiastic.

nytimes.comWhen A.I. Comes for Your Purpose in LifeIn the span of just a few months, A.I. systems have solved some of the biggest outstanding problems in math — adding mathematicians to the list of those terrified of losing their job to the technology...

My night-time skunk companion, Stink Floyd, is not sure what to make of the newly-appeared bags of mulch. [90 seconds, no sound]

We don't appreciate YouTube enough for how it's made it possible BOTH for musicians of all kinds -- from all over the world -- to share their talent globally and for music-lovers to find incredible performances in almost any genre. I should probably spend more time listening to YT vs. arguing on SM

I love how the takeaway from the BS community is, once again "[such and such] TECH BAD!" vs. "Companies should be responsible for better security and be penalized for breaches"

Elizabeth Sandifer@eruditorumpress.com

The same bozos who thought putting refrigerators on the Internet was a good idea are now trying to sell you on AI assistants.

Is a web browser + screen on a fridge kinda stupid? Yes, yes it admittedly is. But I can think of many reasons why it's useful to have appliances connected to the internet (WITH, again, good security). And, related, I find my Smart Home stuff super-useful!

Adam Lasnik reposted

Replying to @lastpositivist.bsky.social

Whatever clout you think you get for always asserting the maximally cynical "it's always been bad and I've always known as much" perspective is worthless and actually making you much stupider as a person and more useless as a political agent.

Adam Lasnik reposted

From The Lancet: in an urgent care setting, the advice of the obsolete Gemini 2.5 Pro & Gemini 2.5 Flash (without access to patient medical records) were rated of similar quality to doctors by other physicians. There was no safety issues spotted. Models have gotten significantly better since.

Adam Lasnik reposted

There's growing experimental evidence¹˒²˒³˒⁴ that LLMs are unusually good at talking people out of entrenched false beliefs and conspiracy theories. ¹ pubmed.ncbi.nlm.nih.gov/39264999/ ² pubmed.ncbi.nlm.nih.gov/41179846/ ³ sciety.org/articles/act... ⁴ connecticut.adl.org/resources/re...

Der Fisher@blackholefun.bsky.social

Unpopular opinion: LLMs are generally deradicalizing people whose beliefs are based on pure propaganda. I've been watching this happen slowly with some people I care about. ChatGPT is one of the reasons they finally accept that the NRA, OAN, and NewsMax are lying to them.

This still remains one of my fav quotes from one of my fav shows. Well, fork. Thanks to the crappy character limit here, I can't share the full quote in a single post. Guess I'll post it elsewhere. FFS, I can't be the only one that's REPEATEDLY frustrated by this anemic character limit, am I?!?

Ranting again re character limits: - Facebook: 63K - Reddit: 40K - Twitter: 280, but 25K when paid - Threads: 500 (10K with attachment) - Discord: 2K - Mastodon: 500 ... - Bluesky: 300 😡🤬😖 2/2

I might quit my gym classes over... THE "MUSIC" 🤮 I told the instructor that if my ears could cry, the entire floor would be flooded. She noted sadly "I don't pick the music." Clearly the folks who license it pay by the chord change ($0).

I dunno how much more of this un-melodic thump thump monotone crap I can take, because I want to hurl the weights rather than lift them.

Adam Lasnik reposted

Replying to @largotg.bsky.social

I mean, if my friend was constantly making siri say it wanted to kill itself, I would probably worry for that friend and make sure they were doing ok. I wouldn’t be concerned for Siri.

I do wish Suno was a bit, uh, less conservative re chord progressions :\. (but I found the vocal pretty damn impressive!)

I don't think we've seen someone with the combo of such evilness and imbecility in our lifetime, though Trump's admittedly got a lot of competition with his lackeys and appointed cronies.

Justin Hendrix@justinhendrix.bsky.social

United States President Donald Trump issues executive order instructing federal government agencies to "use the terms 'Super Intelligence' and 'SI' in place of 'Artificial Intelligence' and 'AI' and will not acknowledge the usage of 'Artificial Intelligence' and 'AI' in any applicable setting."

Continuing to be amazed, grateful, and concerned about AI progress... this time, re marketing'ish videos. I made this with Claude (which leveraged Gemini & Suno) in about 10 minutes of work on my end, semi-ironically about how difficult it apparently is to make genuinely high-quality AI films.

View full thread

But small businesses / startups will likely just use Claude w/ higher effort. So on one hand, this is likely to negatively impact Marketing employment. On the other hand, it offers inexpensive & speedy marketing options (or at least drafts!) to any non-profit or even kid!

Will this be a net destroyer of jobs or a net creator of opps and employment in other areas? Or will it make entrepreneurship easier and potentially more profitable? Maybe a combo? Don't know! But as I've said in other threads, this is reality, and it behooves us all to experiment and learn.

Continuing to be amazed, grateful, and concerned about AI progress... this time, re marketing'ish videos. I made this with Claude (which leveraged Gemini & Suno) in about 10 minutes of work on my end, semi-ironically about how difficult it apparently is to make genuinely high-quality AI films.

IMHO, this level of quality is -- for many purposes -- likely "good enough," with the time & money costs essentially insignificant. And that's with ~1 core prompt! Sure, rich firms will keep paying boutique agencies or talented individuals.