<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Hacker News: ball_of_lint</title><link>https://news.ycombinator.com/user?id=ball_of_lint</link><description>Hacker News RSS</description><docs>https://hnrss.org/</docs><generator>hnrss v2.1.1</generator><lastBuildDate>Wed, 18 Nov 2026 05:12:17 +0000</lastBuildDate><atom:link href="https://hnrss.org/user?id=ball_of_lint" rel="self" type="application/rss+xml"></atom:link><item><title><![CDATA[New comment by ball_of_lint in "Running coding agents without handing them my credentials"]]></title><description><![CDATA[
<p><a href="https://imbue.com/product/latchkey" rel="nofollow">https://imbue.com/product/latchkey</a> is similar/related. Maybe a bit more focused on giving agents access to external services vs your own laptop.</p>
]]></description><pubDate>Tue, 17 Nov 2026 08:10:29 +0000</pubDate><link>https://news.ycombinator.com/item?id=50528765</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=50528765</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50528765</guid></item><item><title><![CDATA[New comment by ball_of_lint in "Conductor is joining Meta"]]></title><description><![CDATA[
<p><a href="https://imbue.com/product/sculptor" rel="nofollow">https://imbue.com/product/sculptor</a></p>
]]></description><pubDate>Wed, 04 Nov 2026 23:17:51 +0000</pubDate><link>https://news.ycombinator.com/item?id=50372793</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=50372793</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50372793</guid></item><item><title><![CDATA[New comment by ball_of_lint in "DeepInfer: High-performance inference framework for DeepSeek models"]]></title><description><![CDATA[
<p>I think there is also an unlikely but reasonable concern that there could be exploits intentionally trained into the model itself.<p><a href="https://news.ycombinator.com/item?id=50157302">https://news.ycombinator.com/item?id=50157302</a></p>
]]></description><pubDate>Tue, 20 Oct 2026 20:15:50 +0000</pubDate><link>https://news.ycombinator.com/item?id=50172042</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=50172042</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50172042</guid></item><item><title><![CDATA[New comment by ball_of_lint in "'500-hour' Order of the Sinking Star is as complex and exhausting as it sounds"]]></title><description><![CDATA[
<p>Order has been great for me so far. I can understand it being difficult/tiring for some, but for those who don't find it so the puzzle design is simply amazing.<p>I'm about 20h in and agree the story is pretty meh, but that's no surprise coming from Blow.</p>
]]></description><pubDate>Fri, 16 Oct 2026 03:59:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=50113908</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=50113908</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50113908</guid></item><item><title><![CDATA[New comment by ball_of_lint in "Someone on Torrent Sites Kept Fucking with My Sonarr"]]></title><description><![CDATA[
<p>> malware<p>> you let it download<p>Sounds like a terrible strategy to me</p>
]]></description><pubDate>Thu, 15 Oct 2026 22:04:40 +0000</pubDate><link>https://news.ycombinator.com/item?id=50111031</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=50111031</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50111031</guid></item><item><title><![CDATA[New comment by ball_of_lint in "A statement on the Tor Project's relationship with Mullvad"]]></title><description><![CDATA[
<p>From the article:<p>> Members of the Tor community have asked us to explain how continued collaboration with Mullvad fits with the Tor Project's human rights mission.<p>> Our decision followed a thorough due diligence process that involved extensive discussions with community members, staff, and the board, a staff survey, and detailed financial analysis and scenario planning.<p>Tor is not a democracy and doesn't purport to be. The article outlines the due diligence they have already done, and they're writing this as notice to their community, in response to questions from it.</p>
]]></description><pubDate>Fri, 09 Oct 2026 21:16:53 +0000</pubDate><link>https://news.ycombinator.com/item?id=50026653</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=50026653</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50026653</guid></item><item><title><![CDATA[New comment by ball_of_lint in "A statement on the Tor Project's relationship with Mullvad"]]></title><description><![CDATA[
<p>Tor isn't saying 'This speech should not be allowed', they're saying 'This speech is speech we disagree with, and don't want associated with TOR'.</p>
]]></description><pubDate>Fri, 09 Oct 2026 20:52:19 +0000</pubDate><link>https://news.ycombinator.com/item?id=50026437</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=50026437</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50026437</guid></item><item><title><![CDATA[New comment by ball_of_lint in "OpenAI fires three safety researchers for "mishandling research information""]]></title><description><![CDATA[
<p>No one has ruled out a suit.</p>
]]></description><pubDate>Fri, 09 Oct 2026 17:34:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=50023979</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=50023979</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50023979</guid></item><item><title><![CDATA[New comment by ball_of_lint in "The Mathocalypse"]]></title><description><![CDATA[
<p>You also have to count the amount of tokens and human time spent afterward to verify these proofs. There's already been some papers retracted and it seems likely to me that there will be many more as people continue digging into the more dubious ones.</p>
]]></description><pubDate>Thu, 08 Oct 2026 17:42:22 +0000</pubDate><link>https://news.ycombinator.com/item?id=50009119</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=50009119</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=50009119</guid></item><item><title><![CDATA[New comment by ball_of_lint in "Why Common Lisp is now the best programming language"]]></title><description><![CDATA[
<p>Lots of bad reasoning in this article.<p>So long as your feedback loop isn't slow enough to be taking you out of flow, it's fast enough. Having it be a few seconds vs a few hundred milliseconds mostly doesn't matter for a human, and will matter even less for an LLM.<p>On macros and DSLs, yes they're cool and even useful sometimes, but most of the software industry is working quite happily without them. And LLMs aren't going to change that because they are best when there's a lot of relevant patterns in their training data. That ends up being a far more important factor in their effectiveness than whether the language itself is token-efficient. It's much easier for an LLM to reason through how to do XYZ in Python where it already knows all the semantics and syntax than it is for it to do it in your DSL it's never seen before. To be clear, it can probably do both but will make mistakes an order of magnitude more in the DSL, and that's what will matter most for the LLM iteration time.</p>
]]></description><pubDate>Tue, 06 Oct 2026 06:19:01 +0000</pubDate><link>https://news.ycombinator.com/item?id=49974895</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=49974895</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49974895</guid></item><item><title><![CDATA[New comment by ball_of_lint in "Religious scholars met with Anthropic"]]></title><description><![CDATA[
<p>Why that particular substrate instead of a slightly narrower one? Like that would include you and your family but not animals like octopi, dolphins, etc?</p>
]]></description><pubDate>Sun, 04 Oct 2026 16:51:35 +0000</pubDate><link>https://news.ycombinator.com/item?id=49955563</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=49955563</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49955563</guid></item><item><title><![CDATA[New comment by ball_of_lint in "I quit OpenAI because its culture is broken"]]></title><description><![CDATA[
<p>That stupidity is happening? Even after the Huggingface hack, frontier labs are using internal models to further their research. i.e. RSI is happening now and we're facilitating it.<p>To make the the argument that P(doom) is real and worth considering, you don't have to say that a fast takeoff is very likely. You don't have to make the argument that RSI to infinity is going to be super cheap, barely even an inconvenience. You just have to show that it has some non-zero probability. And then you start weighing probability of extinction versus finite, mild discomfort now. I don't think anyone is arguing we should let people starve to slow AI progress, instead just some capitalists make less money soon.<p>There are arguments against taking P(doom) seriously that lie in something like having exponential (instead of hyperbolic) time discounting of utility (so you can take the entire future of humanity as a finite utility value). Or in saying that P(doom) is zero or infinitesimal.<p>"Build it and Pray" is the default strategy that we're in, but it doesn't have to be the strategy we choose, and it's unlikely to be the best strategy.</p>
]]></description><pubDate>Sun, 04 Oct 2026 05:47:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=49951037</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=49951037</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49951037</guid></item><item><title><![CDATA[New comment by ball_of_lint in "Ask HN: What are you reading?"]]></title><description><![CDATA[
<p>Big fan of Camus; I found The Fall and The Plague to both be stronger books than The Stranger</p>
]]></description><pubDate>Tue, 29 Sep 2026 22:55:10 +0000</pubDate><link>https://news.ycombinator.com/item?id=49901932</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=49901932</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49901932</guid></item><item><title><![CDATA[New comment by ball_of_lint in "Hijacking the PS5's RTMP stream"]]></title><description><![CDATA[
<p>Can confirm. Have been playing games on Linux for years now and with proton, the vast majority of games just work. And there's protondb to fix up the rest.<p>The one game I couldn't actually get working was Escape from Tarkhov because it has some windows-only anticheat. I.e. the developers chose to make it not work, not an accident.</p>
]]></description><pubDate>Tue, 29 Sep 2026 06:22:18 +0000</pubDate><link>https://news.ycombinator.com/item?id=49888986</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=49888986</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49888986</guid></item><item><title><![CDATA[New comment by ball_of_lint in "There are no "rogue" AI agents"]]></title><description><![CDATA[
<p>Yes, we shouldn't let OpenAI off the hook.<p>But also, these hacks are shots across the bow for AI alignment and safety research. We're fortunate that hasn't been significant damage already. We have to assume that future models will have even greater hacking ability and be closer to having their own desires/goals.<p>So while I agree this language choice is wrong in that it shifts blame away from the company, it is right in that we need to treat this as if these models have their own desires, because we cannot yet determine or set what those are in practice.</p>
]]></description><pubDate>Sun, 27 Sep 2026 17:34:54 +0000</pubDate><link>https://news.ycombinator.com/item?id=49868861</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=49868861</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49868861</guid></item><item><title><![CDATA[New comment by ball_of_lint in "An agent used DNS to reach an external chatbot"]]></title><description><![CDATA[
<p>It's what we know mattered for this model. It doesn't tell us about future models; we already know this strategy is unsound, so it is silly to rely on it.</p>
]]></description><pubDate>Sun, 27 Sep 2026 17:21:02 +0000</pubDate><link>https://news.ycombinator.com/item?id=49868719</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=49868719</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49868719</guid></item><item><title><![CDATA[New comment by ball_of_lint in "An agent used DNS to reach an external chatbot"]]></title><description><![CDATA[
<p>If this thing is too dangerous to be allowed to access the Internet, should it be created at all?</p>
]]></description><pubDate>Sun, 27 Sep 2026 16:30:37 +0000</pubDate><link>https://news.ycombinator.com/item?id=49868215</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=49868215</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49868215</guid></item><item><title><![CDATA[New comment by ball_of_lint in "An agent used DNS to reach an external chatbot"]]></title><description><![CDATA[
<p>Two things can be true. Yes this appears to be negligence on the part of OpenAI.<p>However, making a secure 'sandbox' is quite hard. There's a huge variety of exploits that exist today, including many we don't know about. Strong models have already shown a capability of finding and using such bugs.<p>Even one of the strongest boxes we can imagine, literally just a text interface a human can read, has been repeatedly shown to allow unfriendly AI to escape containment: <a href="https://www.lesswrong.com/w/ai-boxing-containment" rel="nofollow">https://www.lesswrong.com/w/ai-boxing-containment</a></p>
]]></description><pubDate>Sun, 27 Sep 2026 16:24:28 +0000</pubDate><link>https://news.ycombinator.com/item?id=49868142</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=49868142</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49868142</guid></item><item><title><![CDATA[New comment by ball_of_lint in "Yes, Claude can do nine loops"]]></title><description><![CDATA[
<p>I think it's mostly just what it is on it's face; You're telling it to be independent for a while. The default prompts start in an interactive mode where it checks in and verifies requirements a lot.</p>
]]></description><pubDate>Fri, 25 Sep 2026 19:23:47 +0000</pubDate><link>https://news.ycombinator.com/item?id=49848795</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=49848795</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49848795</guid></item><item><title><![CDATA[New comment by ball_of_lint in "Sex, AI, and the Apocalypse"]]></title><description><![CDATA[
<p>I would expect rationalists to be overjoyed to see unaffiliated and very different thinkers taking the topic of AI safety seriously. The flaws you're pointing at are probably accurate; I don't think that implies we should be dismissive of their concerns the way the article is.</p>
]]></description><pubDate>Fri, 18 Sep 2026 01:11:12 +0000</pubDate><link>https://news.ycombinator.com/item?id=49749001</link><dc:creator>ball_of_lint</dc:creator><comments>https://news.ycombinator.com/item?id=49749001</comments><guid isPermaLink="false">https://news.ycombinator.com/item?id=49749001</guid></item></channel></rss>