871 points · theanonymousone · comments

# Summary The discussion criticizes Google for fragmenting Android by releasing new APIs exclusively on Pixel devices before AOSP, potentially withholding security patches from other OEMs, and creating barriers for alternative projects like GrapheneOS—seen as a betrayal of Android's open-source promise.

782 points · ethanhawksley · comments

# Summary Commenters largely criticize passkeys as a poorly-designed, supply-side solution that creates new risks (account lockout, device loss, ecosystem lock-in) while solving problems most users don't have, with confusing UX, poor cross-device support, and implementation issues that make them more frustrating than passwords for average people.

728 points · jcbhmr · comments

# Summary Commenters criticize Cloudflare Quick Tunnels as a rebranded 5-year-old product with poor marketing (dismissing it as "vibe-coded"), while debating its security implications due to anonymous access and comparing it unfavorably to alternatives like Tailscale and ngrok. The discussion also highlights existing issues with Cloudflare's tunnel infrastructure, including maintenance problems and high latency variance.

672 points · datadrivenangel · comments

# Summary The HackerNews community views Claude Code's new AGENTS.md support as overdue bare-minimum compliance with cross-platform standards, with widespread frustration that Anthropic was forced into this change by user exodus to competitors rather than adopting it proactively. Many commenters praise the standardization while criticizing Anthropic's broader user-hostile practices, including poor instruction-following and settings handling that override user configurations.

637 points · ilreb · comments

# Summary The discussion expresses skepticism about OpenJev as a meaningful innovation, with commenters questioning how it differs from existing LLM structured output features and noting that the real Jev's value lies in its proprietary model and speed guarantees—not just the API interface. Many also criticize the project's website design and lack of clear explanation of what problem it solves.

462 points · realsarm · comments

# Summary The discussion centers on concerns that the US military nearly acted on AI-generated false intelligence, highlighting how LLMs systematically produce errors ("hallucinations") that are inherently unreliable for critical military decisions—a danger compounded by historical patterns of politically-motivated intelligence failures and insufficient human oversight.

246 points · ereiamjh · comments

# Summary The discussion reveals a fundamental tension: AI-generated posters can be "good enough" for budget-constrained users, but they suffer from predictable mediocrity—favoring clichéd tropes, repetitive layouts, and surface-level design—which people find frustrating precisely because they're everywhere, signaling laziness rather than genuine creative effort.

52 points · hmokiguess · comments

The discussion reflects skepticism about AI protests in Montreal, with commenters questioning the accuracy of protesters' claims (water usage, EMF concerns), dismissing them as ineffective activism, and noting that AI's proliferation is inevitable regardless of opposition.

109 points · rbanffy · comments

# Summary The discussion reveals that AI chatbots are persuasive not because they're intelligent, but because they're patient, non-confrontational, lack ego investment in debates, and exploit human cognitive biases—while often prioritizing rhetorical convincingness over factual accuracy, which poses serious risks for misinformation and democratic discourse.

145 points · nsoonhui · comments

The discussion largely questions whether GPT-6 actually "solved" the cipher or merely retrieved an existing solution from its training data, with many commenters noting that simple cryptanalysis or dictionary attacks could have worked, and arguing the headline should say the LLM "helped" rather than "solved" the problem.

19 points · Muhammad523 · comments

# Summary AI posts dominate HN because there's genuine widespread interest across enthusiasts, skeptics, and the fatigued—combined with industry hype, possible platform manipulation, and the natural tendency for tech communities to fixate on cutting-edge trends. HN functions as a hype machine that cycles through hot topics, and AI is simply the current wave.

66 points · slater · comments

The discussion centers on whether Y Combinator's PAC supporting Republicans is controversial, with commenters debating the motivations behind tech leaders funding GOP candidates—ranging from claims of balance and single-issue politics to accusations that such support enables authoritarianism and undermines democracy.

320 points · LiamPowell · comments

212 points · karakoram · comments

177 points · axiologist · comments

141 points · num42 · comments

208 points · rbanffy · comments

147 points · iamnothere · comments

65 points · vismit2000 · comments

141 points · CameronBanga · comments

245 points · m-hodges · comments

291 points · emigre · comments

311 points · throw7 · comments

161 points · ibobev · comments

26 points · nadis · comments

33 points · hentrep · comments

21 points · SanjayMehta · comments

195 points · synack · comments

14 points · tommymcm · comments

55 points · asplake · comments

94 points · lioeters · comments

93 points · wittydeveloper · comments

200 points · HenryNdubuaku · comments

104 points · jegp · comments

210 points · mmh0000 · comments

19 points · gherlein · comments

38 points · berkeleyjunk · comments

35 points · Bluestein · comments

25 points · amirmc · comments

10 points · matt_oriordan · comments

30 points · saikatsg · comments

10 points · ejcho · comments

136 points · maxall4 · comments

36 points · jimjag · comments

17 points · toomuchtodo · comments

26 points · n_plus_1_acc · comments

9 points · mdp2021 · comments

34 points · harisingh1612 · comments

10 points · mdp2021 · comments

7 points · curiousgal · comments