Tech Policy Press

The US has executed people in international waters over the claim of fentanyl being trafficked into the country. Is Insta and TikTok as addictive as fentanyl? If so, does it warrant a similar response? I think a cheeseburger is not an equivalent analogy. Singapore also executes drug traffickers, for what it’s worth. https://www.techpolicy.press/is-tiktok-digital-fentanyl/ https://www.foxnews.com/media/tiktok-is-chinas-digital-fentanyl-fcc-co…

But he said he was doing it just for love!! [1] 1: https://www.techpolicy.press/transcript-senate-judiciary-subcommittee-…

This is sorta true. But it turns out there that there are always targets, even if they aren't explicit (is there a name for this law?). So whether you have a recommendation system or a chronological feed or whatever other way you want to display info—there is always that implicit Goodhartish: measure → target → behavior change. You can't escape it. The only question is how you want to harness it—how you can bring out the best in people (with their consent ideally!) and mitigate the negative impacts of Goodharting. (Shameless plug, I'm working on this, e.g. https://techpolicy.press/can-algorithmic-recommendation-systems-be-goo… )

Most of the people working on AI, and even those on the specific sub-domain of AI where Roko's basilisk was coined which isn't the majority of the field by a long shot, have been rolling their eyes at Roko's basilisk since the moment it was coined. Even a brief moment of thought should reveal that, even if you think the scenario likely, there are an infinite number of potential equivalent basilisks and you'd need to pick the correct one. I'm less worried about Roko's basilisk*, and rather more worried about the people who say this: I think you have said in fact, and I'm gonna quote, development of superhuman machine intelligence is probably the greatest threat to the continued existence of humanity. End quote. You may have had in mind the effect on, on jobs, which is really my biggest nightmare in the long term. - https://www.techpolicy.press/transcript-senate-judiciary-subcommittee-… Because this is clearly not taking the words themselves at face value; either you should dig in and say "so why should we allow it at all then?" or you should dismiss it as "I think you're making stuff up, why should we believe you about anything?", but not misread such a blunt statement. (If you follow the link, Altman's response is… not one I find satisfying). * despite the people who do take it seriously, as such personalities have always been around and seldom cause big issues by themselves; only if AI gets competent enough to help them do this do they…

I don't know what to tell you other than this is well-established [1][2][3][4]. Also, what exactly is a "well run utility"? IMHO all utilities should be municipality or state owned. All privatization does is transfer wealth from the government and the not-wealthy to the already-wealthy. I suspect you might not agree however. There are caps on how much utilities can charge but they're allowed to absorb capex (eg by building new transmission lines to a new DC) and if the utility has to buy electricity on the spot market because of increased demand (as was the case with crypto mining in upstate New York), then that raises the per-kWh cost of electricity for everyone, which is great in a cost-plus model. We've seen this exact thing with healthcare insurance premiums. By law, a certain percentage of premiums has to be spent on giving care. Sounds good, right? So how do you, as an insurance company get around that? You push for higher premiums because that same percentage of profit now means more money. And how do you increase healthcare spend to keep that percentage intact? By spending with providers you also own. [1]: https://www.bloomberg.com/graphics/2025-ai-data-centers-electricity-pr… [2]: https://energyathaas.wordpress.com/2025/09/29/what-will-data-centers-d… [3]: https://www.techpolicy.press/how-your-utility-bills-are-subsidizing-po… [4]: https://substack.perfectunion.us/p/how-data-centers-are-driving-up-you…

In a Supreme Court case the burden of proof is on the petitioner, and it's the duty of the Court to critically question that evidence and proof. See, for instance, comparable coverage on the Roe vs Wade case, where you will see similar grilling. [1] Something quite important is how the judges responded to the claims and the follow ups. For instance in the transcript [2] follow the dialogue from the quotes you gave versus the grilling of Prelogar, representing the government. She was, at times, being overtly mocked with quotes from judges like, "That's your best argument is that the average American won't be able to figure out that the cat feed he's getting on TikTok could be manipulated, even though there's a disclosure saying it could be manipulated?" Prelogar in general found herself struggling to defend the claim that the attempted ban was based on data access and not content (which would be unconstitutional), why there were no alternatives if the claims were based solely on data access, and the implications for any other foreign company that has access to user data (which is basically all of them). TikTok was met with some tough questioning but generally responded competently. [1] - https://www.scotusblog.com/2021/12/majority-of-court-appears-poised-to… [2] - https://www.techpolicy.press/transcript-us-supreme-court-oral-argument…

Right, I pointed out the competition to show that the donors are suffering actual damages from the restructuring. I don't think any of the critics here seriously expect the PBC model to fulfill the nonprofit's stated mission in any case. This is not just an issue for the change that OpenAI is contemplating right now, but also the capped for-profit change that happened years ago. If that's found to be improper, I'm curious if that entitles the donors to any kind of compensation. ETA: looking into this, I found the following precedent (https://www.techpolicy.press/questioning-openais-nonprofit-status/). > A historic precedent for this is when Blue Cross' at the time nonprofit health insurers converted into for-profit enterprises. California Blue Cross converted into what's now Anthem. They tried to put a small amount of money into a nonprofit purpose. The California Attorney General intervened and they ultimately paid out about $3 billion into ongoing significant health charitable foundations in California. That's a good model for what might happen here. So my guess is there's no compensation for any of the donors, but OpenAI may in the end be forced to give some money to an open, nonprofit AI research lab (do these exist?). IANAL so that's a low-confidence guess. Still, that makes me so queasy. I would never donate to a YC-backed nonprofit if this is how it can go, and I say that as a YC alum.

“are”? Source? I've seen an absolute flood of “BREAKING:” claims online recently which are quotes from 3 years ago where all of their context is removed, or that quotes from 3 years ago are “because of GLM 5.2”, etc. https://www.techpolicy.press/transcript-senate-hearing-on-principles-f…

Two thirds of Europeans want this - https://www.techpolicy.press/almost-two-thirds-of-europeans-back-repla… > The figures were almost universal across all categories: 62 percent of those surveyed across the five European countries said they favored or had considered replacing US data storage and payment services, while 59 percent of respondents said they would back a change from American video-conferencing companies like Zoom. (Technically only five countries in the EU in this survey, but the five most populous countries, and presumably other countries generally agree)

From the thread, "If you want an explainer on why the EU’s DSA Fine Against X is Not About Speech or Censorship read this article:" https://www.techpolicy.press/the-eus-fine-against-x-is-not-about-speec…

Republicans and Elon Musk have become very skilled at exerting political influence in the US [1] and Europe [2] through social media in ways the public isn't really aware of. Is this really that far from the goal of Cambridge Analytica of influencing elections without people's knowledge? Is it fine for large online platforms to influence election outcomes? Why wouldn't an online platform be used to this end if that's beneficial for it and there is no regulation discouraging it? [1] https://www.techpolicy.press/x-polls-skew-political-realitie... [2] https://zenodo.org/records/14880275

Seems not to if you're working at an AI image generation company. https://cyber.fsi.stanford.edu/news/investigation-finds-ai-image-gener… https://www.techpolicy.press/laion5b-stable-diffusion-and-the-original… Someone somehow downloaded the images in LAION 5B to do the actual training, and we know that thousands of these images contained illegal content. Where's the strict liability? Everyone who ever downloaded and ran Stable Diffusion 1.5, or even Lora's from it, could in some way be held "strictly liable" for the fact that you are simply one prompt away...

If you really haven't read a single argument about it then you're deliberately blocking them out, because it just takes a couple minutes of searching. https://www.arl.org/blog/training-generative-ai-models-on-copyrighted-… https://hls.harvard.edu/today/does-chatgpt-violate-new-york-times-copy… https://www.bakerdonelson.com/artificial-intelligence-and-copyright-la… https://www.techpolicy.press/to-support-ai-defend-the-open-internet-an…

I recall he made changes at the beginning of the year specifically because "it's time to get back to our roots around free expression." Full transcript of his remarks can be found here: https://www.techpolicy.press/transcript-mark-zuckerberg-announces-majo…

& the US system routes abuses through the private sector: https://www.techpolicy.press/the-us-government-buys-data-for-surveilla… These things aren't a one-and-done matter of legislation or constitutions, they rely on constant pressure on every case. (and I explicitly said the EU system does not guarantee privacy against law enforcement! Because total privacy for crime is very unpopular and politically unsustainable)

The Scaling Myth: One argument proposes that many problems with AI will be fixed with more data or better training. ... The idea that this is a path to a “general” intelligence — a system that can perform nearly any task at or beyond a human level — is still highly speculative. https://www.techpolicy.press/challenging-the-myths-of-generative-ai/

More to the point, when Google sought approval to buy DoubleClick, they testified before congress that they would not merge information gleaned from your use of Google services with your advertising profile. If their CEO's congressional testimony on this point isn't considered binding at Google, verbiage on their website certainly isn't. > To assuage concerns, Google told Congress and the FTC that it would not combine the user data it got from assets in search, e-mail, or GPS maps with information from DoubleClick about which consumers visited which publications. And so, the acquisition was greenlighted. Ten years later, though, Google did not hesitate to break its promise. https://www.techpolicy.press/how-us-history-and-googles-own-behavior-j…

Too bad not disclosing that you always intended to convert the non-profit into a for-profit during your testimony while numerous senators congratulate you about your non-profit values isn't problematic. https://www.techpolicy.press/transcript-senate-judiciary-subcommittee-…

Profiling people for advertising, building shadow profiles, buying 3rd party user data, tracking people across the web, devices and geographically, psychological experiments, optimizing for engagement, limiting post reach without pay... generally prioritizing profit over people. https://www.techpolicy.press/is-it-ethical-to-work-at-facebook/

"It's not perpetuating stereotypes".. ok https://techpolicy.press/researchers-find-stable-diffusion-amplifies-s…

Kudos for nice work! What did you make of Mudge's report regarding resiliency of data-centers? > Insufficient data center redundancy, 59 without a plan to cold-boot or recover from even minor overlapping data center failure, raising the risk of a brief outage to that of a catastrophic and existential risk for Twitter's survival. - https://techpolicy.press/wp-content/uploads/2022/08/whistleblower_disc…

> - This Congress is ill-equipped to understand tech, much less hold it accountable. As long as the people are happy, Congress is happy. There's an article I was introduced to yesterday: Do We Need a New Digital Regulatory Agency in the U.S.? It argues that it it is the agencies and the experts within the agencies that need to become more technologically literate to be able to advise creation and implement the laws that have tech impacts. Congress isn't supposed to be experts on subjects, they're supposed to be the representatives of their people with occasional domain knowledge in certain areas of importance to their constituents. We can't (and shouldn't) expect every member of congress to be an IT expert. https://techpolicy.press/do-we-need-a-new-digital-regulatory-agency-in… ( https://news.ycombinator.com/item?id=32555365 )

> It’s clear there are forces pushing censorship in the continent of Europe. I don't believe that the author Mike Masnick is denying the existence of pro-censorship forces within the EU, and Masnick has criticized the censorship aspects present in the DSA itself [1] (as distinguished from censorship relating to only specific interpretations and implementations of the DSA, as in Thierry Breton's case). > Trying to claim there is “no EU Internet Censorship” is a blatant lie. The claim was not that there is literally no EU internet censorship, but that the US federal government under the Trump administration has been claiming instances of EU internet censorship while thus far failing to identify any such instances even though the administration controlled which criteria its investigators used. As I interpret it, the purpose of TFA is to point out that: - The US federal government has been using instances of non-censorship as evidence of censorship (e.g. TFA cites [2], which cites [3]). - The US federal government has censored speech in ways that the DSA could do but hasn't yet done, and the US justified its own censorship by pointing to supposedly-already-happening DSA censorship that hasn't happened. - The EU "removed" Thierry Breton (actually, he resigned), the current US administration has yet to punish one of its own censors. (This argument is weak to the point of uselessness because Breton's resignation might be unrelated to his censorship, but it's an argument that…

> that Google spends >$1B/year to develop. Let's see... https://www.techpolicy.press/the-true-cost-of-browser-innovation-why-c… * Most of the personnel involved in developing web technologies are engineers, but they also include product managers, sales, marketing, legal, customer support, and other functions. * Given the complexity of Chrome and web technologies, the engineering teams skew towards higher levels of seniority. Assume that Staff Software Engineer is the most common engineering level represented across the web technologies teams, which is towards the more senior end of Google’s software engineering job ladder. * The average base salary for Google employees working on web technologies is $240k and the average annual take-home pay is $500k, including salary, bonuses, and stock payments. These estimates are close to the current average base salary and take-home pay for Google Staff Software Engineers listed on industry salary data sites. * Google has approximately 2000 staff working on web technologies. Using the above assumptions, the estimated personnel cost for web technologies is 2000 * $596k = $1.2B. Of course there are additional costs associated with these businesses. Based on this sketch, it seems fair to assume that Google spends at least $1-2B annually on Chrome, Chromium, and the evolution of the web platform.

> Well this is terrifying: the billionaire-backed Evie magazine—a conservative publication that says IUDs can make you hear voices and drown your children—plans to launch a wearable period-tracker that collects data about your cycle, sleep, diet, and movement. > This true nightmare fuel is a project of Brittany Hugoboom, the conservative influencer who created Evie and a period-tracking app with her husband, Gabriel. It gets worse: the Hugobooms are funded by none other than pronatalist Trump ally, Peter Thiel. > That’s right, the billionaire behind Palantir and the rise of JD Vance is very interested in women’s pregnancies: Thiel bankrolled Hugoboom’s “28” app, which has raised all sorts of questions about data-collection and privacy in post-Roe America. The "pronatalist" description of Thiel had me curious, so I looked that up and he does have a history of funding fertility companies and pushing for higher birthrates: https://www.techpolicy.press/tech-oligarchs-and-the-rise-of-silicon-va… https://x.com/AutismCapital/status/1824669688934363177 > PETER THIEL ON DECLINING BIRTH RATE: > “Once you get an inverted demographic pyramid, at some point do you vote for benefits for the old or for the young? If the people with kids get financially penalized and the people without kids vote benefits for themselves, it will be very difficult to flip back.”

> An easy way to solve for this is customization -- if no two users have the same "algorithm" powering their feed, it becomes hard for anyone to do this, because perhaps one user's algorithm filters out anything tagged with politics, or with a low Flesch–Kincaid score, or non-text posts, etc. The problem, and where I strongly agree with the parent's statement that "I feel like social media has changed human behavior", is that the users themselves seek the engagement. Content creators want feedback about their content. You can codify that as "views", "likes", or whatever, but the whole problem here is fundamentally that most creators try and pursue strategies that increase whichever metric they are tracking to get value out of their posting. I watched Bluesky grow up and become a "real network" and once Bluesky hit a certain scaling point it became the exact same as all the other supposed algorithmic-engagement optimized sites. Posters started posting snippy, sneery comments because it made the Like count go up. > perhaps one user's algorithm filters out anything tagged with politics, or with a low Flesch–Kincaid score, or non-text posts, etc. Zuck talked about how Threads specifically filtered out political content [1] and how that decision was reversed [2]. It turns out that users didn't like filtering out political content even though as most of us know, it tends to turn into dunking competitions online. So I largely agree with what the parent said. The expectations…

> Where have I heard this before? Maybe in EU legislation where this privileged access is now mandated by law for every certified fact checker. Although these aren't conventional fact checkers, they just "flag" content. > Trusted flaggers are public and private actors with privileged access to internet platforms through which they can identify illegal and harmful content. Before DSA, trusted flaggers existed as a voluntary system. https://www.techpolicy.press/building-trust-in-the-digital-service-act… > diverse funding streams They are not independent enough and would develop their own interests. > narrative of persecution that serves fascistic rhetorical goals. Yeah, fascists are known to flag content too. Otherwise this is complete conjecture. > no institution can be trusted No institution, which will always be dependent on government funding and special legislation, can be trusted to moderate internet content. That much is true.

EFF is quite popular among "nerds" but as a source for information and advocacy about electronic privacy I actually prefer EPIC.^1 Perhaps someone reading this can explain why, but I did not see anyone from the EFF at the 8 September 2022 public forum on the FTC's ANPR on Commercial Surveillance and Data Security. EPIC was there providing cogent commentary. The operater of thenexusofprivacy.net was there, too.^2 1. For example, here is a comparison of the ADPPA with the CCPA from the folks at EPIC. https://techpolicy.press/evaluating-the-american-data-privacy-and-prot… 2. This is another topic that may interest HN thread readers. This is Section 18 "Mag-Moss" rulemaking so public input is mandatory. Those who understand the issues should submit comment to the FTC to support the process. The deadline is 21 October. https://www.regulations.gov/comment/FTC-2022-0053-0001 https://news.ycombinator.com/item?id=32947397

> And people are using it for revenge porn? I haven't seen that. https://oag.ca.gov/news/press-releases/attorney-general-bonta-launches… https://www.reuters.com/legal/litigation/grok-says-safeguard-lapses-le… https://www.nytimes.com/2026/01/09/technology/grok-deepfakes-ai-x.html https://www.vogue.com/article/grok-deepfakes-trend-essay https://www.the-independent.com/tech/ai-grok-twitter-fake-images-musk-… https://techpolicy.press/the-policy-implications-of-groks-mass-digital… https://www.rollingstone.com/culture/culture-features/grok-sexualized-… The French raided the X offices in Paris. > https://www.theguardian.com/technology/2026/feb/03/french-headquarters… > It said the alleged offences it was investigating now included complicity in the possession and organised distribution of child abuse images, violation of image rights through sexualised deepfakes, and denial of crimes against humanity.