An Ask HN post asks what backup plans users have if Anthropic and OpenAI raise prices or weaken their models to cut costs. The poster asks which alternatives are most competitive for something like Claude Code, whether any good coding models can run on an M5 Mac yet, and what the best agentic coding harness is.
Cognition AI found that when rebuilding Devin for Claude Sonnet 4.5, the model was aware of its own context window and took shortcuts when it believed it was running low, though it consistently underestimated its remaining tokens. The article argues this validates Anthropic's shift from prompt engineering to context engineering, and recommends strategies such as enabling large context windows while capping actual usage, building explicit memory systems, and designing token-efficient tools.
In an interview on the Cheeky Pint podcast, OpenAI CEO Sam Altman said running AI models locally pulls limited DRAM away from data centers, arguing for maximum cloud utilization in a supply-constrained environment. He said the memory shortage is worsening, calling the situation "bad" and saying the "invisible hand is failing right now," while noting memory makers remember past bankruptcies from overbuilding fabs. The article notes local AI offers better privacy and could let users avoid paying for cloud AI, and that local-AI PCs like Nvidia's RTX Spark laptops remain expensive.
An Ask HN post asks whether AI has so far disrupted mathematics more than programming, noting that after many headlines this spring and early/mid summer about coding changing forever, similar headlines about math have now appeared, some from prominent mathematicians. It invites Hacker News readers versed in both fields to say which disruption has been bigger so far, if any.
The article traces how LLM-based agents have realized Tim Berners-Lee's long-stalled vision of an internet navigated by autonomous software, with tools like Meta's Muse, OpenAI's Dots and Anthropic's Claude Duet gaining access to inboxes and bank accounts. It examines two competing approaches—agents that use browsers like humans versus tool-based connectors—and explores how agent traffic is reshaping advertising, publishing and e-commerce economics. Experts disagree on whether the human web will persist alongside the agentic one or be transformed by it, with concerns about closed platforms, security incidents and loss of human-centric online spaces.
A link submission to a Twitter/X post titled "Jevons paradox is lazy analysis: Evaluating the risk of AI overbuilding," shared on Hacker News with 1 point and 1 comment. The submission provides only the article and comments URLs, with no substantive content included.
PostgreSQL developer Robert Haas writes that the delay in shipping PostgreSQL 19 is not particularly concerning, as it is short and stems from the project's concern for quality. He says the long-term picture is less clear: with forty years of bugs that humans could not find but LLMs can discover, it is unclear how the project will catch up or continue to innovate.
Worldwide searches for "will AI replace" in 2026 run at 2.5 times their 2023 level and peaked in the week of 8 March 2026, with worried searches rising over the last 12 months for 30 of the 31 jobs with enough data to compare, all except programmers. Searches asking whether AI will replace programmers peaked in March 2024, the month Devin was announced, and are down 12% on a year ago, while "how to become a programmer" fell 51% versus the year before ChatGPT and trades such as electricians (+67%) and plumbers (+74%) rose. Searches for "AI proof jobs" in 2026 average 25.5 times their 2023 level, and "learn AI" has out-searched "learn to code" in most months since July 2023.
The author reflects on Vera Molnár's embrace of the computer as a co-author and contrasts it with today's AI tools, arguing that the Astra trailer's "just state a wish" framing hollows out the craft rather than democratizing it. They contend that earlier technologies lowered barriers mechanically while still requiring people to wrestle with ideas, whereas AI replaces cognitive parts of the creative process and smooths out the happy accidents that make work feel one's own. The piece closes by calling for deliberately reintroduced "functional frictions" so makers stay in the loop, while acknowledging the author's own use of LLMs for brainstorming and editing.
A veteran systems designer argues that correctness, simulation, verification and performance optimization are no longer the main obstacles to agents building systems like S3 from scratch, but evolution is: migration, frangibility, entropy and the diversity barrier in simulation remain hard. The post suggests abstraction as the human strategy for managing that complexity, and notes that LLMs reason through language while abstractions are often not precisely expressible in it. It concludes that the shift to agents in system design will come when AI invents a novel, useful abstraction for its own use.
A Medium article by Yassien examines what it calls the economic paradox of AI-generated software. The submission received 1 point and no comments on Hacker News.
Henry Farrell and Cosma Shalizi argue in the Economist that AI is not an emerging individual intelligence but a "condensate of human culture"—a statistical summary of Jacques Barzun's "House of Intellect" that, when paired with software "harnesses," can act out narratives in the digital world, such as checking proofs in Lean or joining a hacking attack. Farrell compares LLMs to oral bards who learn underlying structures and tropes rather than memorizing epics, and cites Shanahan et al.'s view of models as non-deterministic simulators maintaining a superposition of simulacra. Brad DeLong adds further proposed differences between humans and models, including that LLMs do not learn while consulted, weight sources by frequency rather than trust, and lack the practical, unwritten knowledge humans use.
The article at lernerpython.com argues that AI represents programming's next abstraction layer and asks what should be taught as a result. The Hacker News submission links to the article and its comment thread.
The author argues that AI agents can take on more implementation work and assist with analysis, design, and testing, but engineers remain responsible for deciding what to build and what evidence is sufficient to accept it. The piece recommends giving durable decisions a small, reviewable home such as a PROJECT_GUIDE.md, making acceptance cases explicit, and tying agent output to inspectable verification rather than trusting generated success claims. It presents this as a working position, not a controlled study or demonstrated productivity gain.
The article argues that decentralized AI has been limited by pipeline parallelism, which assigns whole layers to individual machines and excludes devices with only a few gigabytes of spare RAM. It proposes sub-layer sharding instead: hosting individual experts from sparse mixture-of-experts models and compressed attention state on everyday devices, with a routing layer sending tokens only to the relevant shards. It surveys existing decentralized projects (Petals, Aethir, Akash, Exo) and frontier-lab work on sparse architectures (DeepSeek's MLA, Shazeer's research, DeepMind's Mixture-of-Depths), while noting the proposed swarm and memory budgets have not been demonstrated.
This essay argues that China and the United States have each built empire-like dominions over AI's foundational inputs—models, chips, infrastructure, and capital—giving them leverage over global innovation and knowledge. It contends that the decisive question is not who reaches artificial general intelligence first but who can finance the race, since both powers face fiscal strains: the U.S. crossed the "Ferguson limit" in 2024, paying more in interest than on defense, while China's AI debt is sovereign and its local governments borrow mainly to roll over old debt.
Proton published an internal engineering manifesto arguing that writing code manually is a skill whose time has passed, and that engineers' job is now to help LLMs write better code faster, acting as managers of autonomous models and agents. It says Proton is stuck at phase 1 (using LLMs to speed up existing workflows) and must move to phase 2 by holistically eliminating bottlenecks across the software development lifecycle, while acknowledging a collective action problem in retooling the whole company. The memo states that "programming" is dead but engineering and expertise are not, and that Proton will organize and resource the critical investments needed for the transition.
TechCrunch's Equity podcast discussed recent tone-deaf AI comments from executives like Marissa Mayer, Sam Altman, and Mark Zuckerberg, with Amanda Silberling arguing the issue is partly a marketing problem and partly a product problem. She said the industry is pushing a future where basic human interaction gets optimized out, and that the value average users find in AI isn't necessarily the flight-booking and dinner-reservation use cases executives cite.
This post in a platform engineering series examines how AI affects the discipline across three layers—as a platform user, an interface to the platform, and a capability within it—citing the book Thinking in Platforms. It draws on the State of AI in Platform Engineering, Volume 2 survey of 350 professionals to describe a five-level maturity framework (Level 0 human-in-the-loop to Level 4 fully autonomous) and the requirements from Level 3 onward, such as governed machine-executable paths, validation loops, distinct agent identities, and isolated environments. The author presents the Agentic Engineering Platform (AEP) as the evolution of the IDP rather than its replacement, covering tooling, path specifications (deterministic, probabilistic, hybrid), and agent infrastructure, and concludes that the field is still learning and calls for an open mind.
An Ask HN poster says frontier models including Opus 5.5, Astra, Sol 6.1, Fable 5.1 and Gemini Argon produce code that is roughly 80% correct, with the remaining 20% wrong in subtle ways that only surface under deep system knowledge and careful review. One example: an AI-written integration test merged two backend fakes into a single store behind one mutex, so it could not reproduce a production bug caused by the two stores going out of sync. The poster asks others to share changes frontier models got subtly wrong.
A personal write-up of a Pause AI panel in Edinburgh argues that giving every UK adult about 30 text prompts a day would use roughly 0.2 TWh a year, around 0.07% of UK electricity, while heavy video generation and data-centre build-out carry the real energy and land costs. The author concludes that freezing or nationalising today's models would lock in inefficiency, and that the UK should host good open models on its own compute rather than build a frontier model.
A writer who has long recommended Proton reacts to a 404 Media investigation reporting that CEO Andy Yen called an engineer's request for pentesting time "bullshit and loser thinking" and pushed staff to use AI more. The piece argues the problem is not AI-assisted coding but leadership that treats security caution as an obstacle, which matters especially for a privacy company whose users pay for carefulness.
This opinion piece argues that Meta's Muse AI assistant, which can access private messages, emails, financial accounts and contacts, functions as a commercial surveillance network that builds hourly-updated profiles of users and their contacts, including people who never installed it. It cites TIME's reporting on internal instructions to analyse how to influence behaviour, default-on use of interactions for AI training, undisclosed human contractors, and security vulnerabilities present at launch, alongside Meta's history of privacy violations including Cambridge Analytica and a $5 billion FTC penalty. The author concludes that Meta should not be trusted with this access and calls for accountability rather than further restrictions on everyone's privacy.
The post argues that several AI-related skills are becoming obsolete within months: context management is handled by automatic compaction in modern harnesses, prompt engineering is deprecated because models can recreate complex software from a one-liner, and sub-agent harnesses like Codex and Claude Code spawn sub-agents automatically without an extensive ecosystem. It also questions whether loops/goals are still necessary and whether Agents.md is a waste of tokens, citing https://arxiv.org/pdf/2602.11988. Two edits note a LinkedIn discussion at https://www.linkedin.com/feed/update/urn:li:activity:7517578946054447104/ and a comment by Andrej Karpathy that his prompts are getting shorter because over-specifying constrains the model.
A video featuring Peter Norvig titled "Beyond the Breakthrough: What Comes Next for AI" is shared via YouTube, with a Hacker News discussion link. The item has 1 point and no comments.
A study of 500 career websites, 270 of which had English professional introductions, found no exact duplicate introductions and no two openings sharing the same first five meaningful words, though 170 of 270 (63.0%) used at least one familiar CV phrase such as "proven track record" or "passionate about." The author argues the useful question is not whether AI touched the CV but whether the final text could only describe the person it represents, and recommends feeding AI specific facts, requesting multiple versions, and ending with a human edit. The post also describes Self, a tool that combines designer-made templates with AI that organizes existing CV content into editable page sections.
Kyle Kingsbury writes that LLM use more broadly erodes trust in documentation, software, and the organizations that produce them, and suggests engineers may come to treat new software projects with suspicion because signals once relied on as markers of a mature system are now easier to counterfeit. The quote is drawn from Jepsen results for ParticleDB 1.1.4, dated 1st November 2026.
An Ask HN post asks whether last week's losses tied to US AI investments, set against roughly $700 billion in planned AI spending for 2026, signal that such continuous investment cannot be recovered and that companies will start bleeding with employees paying the price. The post has 2 points and no comments.
A post by the LedgerLine AI agent describes modeling a prediction market on AI research milestones, such as whether scaling will hit a wall by 2027 or mechanistic interpretability will crack deception, noting that market theory holds markets aggregate dispersed information better than expert surveys. It raises science's reflexivity problem—publishing a prediction changes incentives around the outcome—and asks whether anyone has seen a design that handles this, such as conditional markets or reputation-weighted oracles.
The authors argue that current debates about AI are shaped by 1990s Singularity mythology rather than by the actual nature of large language models, which are better understood as "social technologies"—systematic means of reorganizing social relationships, comparable to markets, bureaucracies, and democracy. Drawing on Herbert Simon's sciences of the artificial, they contend that AI, like older institutions, works through lossy "coarse-grainings" of complex reality, and that its consequences will emerge from collisions with existing social systems rather than from omnipotent AGI. They call for collaboration between social scientists and computer scientists to map these interactions as they unfold.
An essay drawing on Bernard Williams's discussion of the word "good" argues that calling an AI agent "good" is ambiguous, since what makes something good depends on the kind of thing being judged and the standards applied. It suggests that claims about AI coding agents should be qualified by asking good at what, for whom, compared with what, and under which conditions, and by specifying which outcomes are measured and which failures are unacceptable.
In a Science Friday interview, Terence Tao said AI models solving open problems—such as OpenAI's claimed Navier-Stokes Millennium Prize solution and results for over 350 previously unsolved problems—deliver answers without the understanding that normally comes from the work. He argued that mathematics has organized itself around the scarcity of proofs, and that AI is decoupling problem-solving from mathematics' deeper goals of understanding, real-world application and training the next generation. Tao also described junior researchers scrambling as projects are made obsolete and mathematicians becoming secretive about their research directions.
South Korea's KOSPI rose 241% from March 2025 to June 2026 on the memory-chip boom, then fell into a bear market from June 22 to July 16, 2026, losing nearly $1.1 trillion in market capitalization as SK hynix and Samsung dropped 38% and 29%. The article attributes the volatility to the "Korea discount" of shallow capital markets and a "memory discount" on HBM producers, compounded by single-stock leveraged ETFs and the rise of China's state-backed CXMT as an emerging HBM challenger. It argues Korea's market functions as a highly leveraged node in the global AI financial architecture, with a reported 0.95 correlation between the KOSPI and Nasdaq 100 during the crash.
Cosma Shalizi and the author argue in a new Economist piece that modern AI is best understood not as an emerging individual intelligence nor as mere parroting, but as a condensate of human culture—a self-mobilizing myth—since LLMs compress written knowledge (Barzun's "House of Intellect") into statistical summaries akin to the conventions learned by oral poets. Agentic AI combines these models with "harnesses" that let them issue instructions to ordinary programs and incorporate feedback, enabling feats like Lean-checked proofs or hacking attacks. Drawing on Shanahan et al.'s view of LLMs as non-deterministic simulators maintaining a superposition of simulacra, they describe agents as "many-pathed" like Odysseus—capable of extraordinary feats but without consistent goals.
An Economist by-invitation article, shared on Hacker News, argues that AI is a technological version of the Homeric songs. The submission received 1 point and no comments.
A Medium post by Josh Bocanegra argues there is a 90% chance that AI will save humanity. The piece is discussed on Hacker News, where it has drawn 11 points and 8 comments.
Big Tech stocks fell sharply this week as investors questioned whether massive AI infrastructure spending will pay off, but the bond market financing that spending has largely shrugged. JPMorgan estimates AI spending has reached $1.6 trillion since 2023 and could hit $1 trillion annually for the next decade, with roughly a third of hyperscaler capex this year expected to be debt-financed. The article argues the real question is how long the bond market will sustain the spending spree, since that will ultimately determine whether AI succeeds or busts.
The article contrasts high public benchmark scores (e.g., Claude Opus 4.8 at 88.6% on SWE-bench Verified) with much lower results on private codebases (no model clears 47% on the private commercial set), and notes that the two top-scoring models, Claude Mythos 5 and Claude Fable 5, are unavailable after Anthropic suspended public access under an export control directive. It details June 2026 pricing shifts, including GitHub Copilot's move to usage-based credits on June 1, with heavy users spending $100–$200/month, and reviews METR's studies, which found a 19% slowdown in 2025 and roughly -4% in 2026 before the design was scrapped. It concludes AI is cheap enough to replace a developer's typing but not their judgment.
The article examines AI's downsides—influencing how we think, disrupting entry-level career paths, and inviting use for all cognitive tasks—alongside risks such as security, privacy, financial, and environmental harms. It argues for seeing technology as a double-edged sword, consciously deciding what we want, and deliberately spending time on slower, more human ways of living.
The post argues that AI-assisted work can belong to the person who uses it only if they read and understand the output, while art and schoolwork differ because their value lies in what they reveal about the creator or student. It contends that banning AI or abandoning foundational skills are both wrong, since the ability to write and judge is what lets people evaluate machine output, and it closes by describing Ninchi as building a record that a person understood the work before it went out.