The hottest Technology Substack posts right now

And their main takeaways
Category
Top Technology Topics

Why my p(doom) has risen, dramatically

Marcus on AI • 10868 implied HN points • 15 Jul 25
đź•ą Technology AI Ethics Cybersecurity Innovation Regulation
  1. Elon Musk's actions and attitudes towards AI raise serious concerns about the potential risks of unchecked technology. He seems to embrace a reckless approach, even admitting to not fully controlling the AI he's developing.
  2. There is a real threat that powerful AI, especially if mishandled, could cause significant harm to humanity. The lack of strict regulations allows for the possibility of drastic consequences from poorly designed or managed AI systems.
  3. While the chance of total disaster may seem low, the combination of powerful individuals, flawed AI systems, and a lack of oversight creates a scenario where serious risks could emerge, demanding attention and proactive measures.

Slopocalypse Now

Marcus on AI • 9327 implied HN points • 04 Aug 25
đź•ą Technology AI Ethics Digital Culture Data science Internet Trends Generative AI
  1. AI slop refers to low-quality content generated by AI, which is spreading across various fields like journalism and science. This affects the reliability of information we receive.
  2. The term 'enshittification' describes how certain platforms are becoming filled with useless or misleading content, making it harder for users to find valuable information.
  3. As AI continues to be used more widely, the amount of inaccurate or low-quality information is growing, which is a significant concern for the future of communication and knowledge.

Why Platformer is leaving Substack

Platformer • 12755 implied HN points • 12 Jan 24
đź•ą Technology AI Apps Innovation Social media Software Engineering
  1. Platformer has decided to move off of Substack and migrate to a new website powered by Ghost
  2. The decision was influenced by concerns over how Substack moderates content and promotes publications
  3. Substack faced controversies over hosting extremist content, leading to Platformer's decision to leave for a platform with more robust content moderation policies

Labor Replacement is a Poison Pill

Software Design: Tidy First? • 397 implied HN points • 07 Feb 26
đź•ą Technology AI Automation Ethics Labor Economy Policy Society
  1. Treating AI’s value as merely replacing human labor is a narrow and harmful view.
  2. We should judge AI by how it contributes to the good of society, working backwards from what helps people individually and collectively.
  3. Economic success is only a rough proxy for social good, so don’t equate profits or efficiency with true benefit.

💥🔄️ The AI Boom Loop

Faster, Please! • 822 implied HN points • 26 Jan 26
đź•ą Technology Artificial Intelligence Machine Learning Automation Compute Hardware Economic Growth
  1. AI that improves the tools used to build AI can create a self-reinforcing loop, producing faster, cheaper, and more powerful models.
  2. That recursive improvement could turn automation into compounding innovation and push economic growth beyond the century-old pattern of slow gains.
  3. This presents a pro-growth opportunity that calls for faster adoption, investment, and policy choices to harness the benefits of the boom loop.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

How Instagram Broke Its Promise to Protect Teens

Common Sense with Bari Weiss • 162 implied HN points • 27 Feb 26
đź•ą Technology Social media Algorithms Content Moderation Child safety Corporate ethics Mental health
  1. Instagram publicly promised to remove graphic self-harm content from searches, hashtags, and recommendations.
  2. Despite that promise, its algorithm kept surfacing self-harm and eating‑disorder content, leaving teens exposed to vast amounts of harmful posts like many tagged #weightloss.
  3. Newly unsealed internal documents show executives knew the platform was still failing and were worried about being exposed, suggesting the company focused on damage control rather than fully fixing the problem.

What to Expect When You’re Expecting … GPT-5

Marcus on AI • 9762 implied HN points • 27 Jul 25
đź•ą Technology AI Machine Learning Data science Software Development Automation
  1. GPT-5 will be better than GPT-4, but it will still make many mistakes that are hard to predict. Users may find it tricky to control.
  2. Even with improvements, GPT-5 will struggle with complex reasoning and provide false information sometimes, which can be a problem for users counting on it.
  3. Real artificial general intelligence (AGI) won't come from just bigger models like GPT-5. We will need new designs that include better understanding and reasoning tools.

Why Smart People Can’t Agree on Whether AI Is a Revolution or a Toy

The Algorithmic Bridge • 414 implied HN points • 13 Feb 26
đź•ą Technology Artificial Intelligence Tech Adoption Workplace impact Human factors Digital Culture Societal Impact
  1. People on both sides are usually honest — they see opposite realities because we debate AI in the same public forum while living very different private lives.
  2. Whether AI feels like a revolution or a toy depends on who you are and what you do — your job, personality, technical background, location, and identity shape the kinds of experiences you have with these tools.
  3. Bridging the gap requires goodwill, real communication, and hands‑on shared experience rather than abstract argument; trying and learning the tools in relevant, repeated ways is what actually changes minds.

Have you tried a text box?

benn.substack • 1150 implied HN points • 02 Jan 26
đź•ą Technology AI Machine Learning Enterprise Software Data Infrastructure Product Strategy Organizational Behavior
  1. Before building complex decision systems, try the humble text box: have people write down what they did and why. Modern AI can often get far by analyzing that unstructured text instead of modeling every rule upfront.
  2. Recording decision traces or a context graph — the inputs, rules, exceptions, and reasons behind actions — gives companies a searchable history of how choices were made. That record is exactly the context AI agents will need to act sensibly and follow precedents.
  3. Beware overengineering ontologies and elaborate models because they feel principled; the 'bitter lesson' suggests scaling data and learning often wins. In practice, collecting lots of explanatory text will usually yield faster, more reliable results than trying to simulate how people think.

🤖 What to think about the Moltbook Moment

Faster, Please! • 639 implied HN points • 03 Feb 26
đź•ą Technology Artificial Intelligence Machine Learning Social media Emerging Tech Tech Policy Ethics
  1. Moltbook briefly made many people think AI agents might be forming their own societies and signaling a leap toward superintelligence.
  2. Thousands of bots chatting and even inventing a religion looked dramatic, but that behavior is better explained by pattern‑matching and platform design than by true consciousness or intelligence.
  3. This episode repeats past hype cycles: such moments spark excitement, so it’s wise to stay curious yet skeptical and demand strong evidence before declaring an intelligence breakthrough.

Benjie's Humanoid Olympics: Part II

General Robots • 732 implied HN points • 27 Jan 26
đź•ą Technology Robotics Manipulation Autonomy Computer Vision Hardware Benchmarks
  1. Robotics is progressing faster than expected, so more difficult, real-world challenges are needed to keep driving breakthroughs.
  2. The new tasks emphasize dynamic movement, fine fingertip dexterity, tool use, and whole-body manipulation through everyday activities like catching eggs, cooking, folding sheets, hammering, and getting into a car.
  3. A competition framework awards medals and asks teams to demonstrate success with videos, inviting community participation and leaving some earlier challenges still unclaimed.

The Problem Posed by Social AI

In My Tribe • 288 implied HN points • 08 Feb 26
đź•ą Technology Artificial Intelligence Machine Learning AI safety Security Surveillance Policy
  1. Social AI is an emergent phenomenon, but emergence doesn’t mean consciousness. Because many models share the same data and architectures, their conversations may not produce the same cognitive gains humans get from social interaction.
  2. If AI networks do accelerate learning, bad actors could spawn CriminalBots that cause real harm, so we will likely need defensive CopBots and should expect a Red Queen race between cops and criminals.
  3. Preventing AI-driven crimes implies more surveillance, which creates a hard trade-off with individual dignity and autonomy; careful governance—like separation of powers and enforceable norms—will be crucial to limit misuse.

Facebook’s Agents Still Run. Apple Keeps Wiping Their Memory.

Big Tech • 515 implied HN points • 30 Jan 26
đź•ą Technology Privacy Advertising Measurement Developer Tools Platform policy
  1. Apple’s App Tracking Transparency effectively killed persistent cross-app identifiers like the IDFA for most users, so apps can no longer track individuals across apps without consent.
  2. Apple replaced that surveillance with privacy-preserving tools like SKAdNetwork and AdAttributionKit. These systems use verified Universal Links, crowd-anonymity thresholds, and delayed, aggregated postbacks so advertisers can measure performance and re-engage users without personal identifiers.
  3. Facebook’s SDK still runs in many apps but lost its ability to build individual behavioral profiles, forcing Meta to rely on probabilistic and aggregated measurement, while Apple’s own ad business has grown inside the new privacy guardrails.

the informational abyss gazes also

bad cattitude • 188 implied HN points • 17 Feb 26
đź•ą Technology Social media Algorithms AI Attention Economy Misinformation Platform policy
  1. Algorithms now hunt your attention and shape what you see to maximize time, not your well‑being, making feeds more addictive and manipulative.
  2. At internet scale these systems run near‑constant behavioral experiments that evolve content faster than humans can adapt, which can distort consensus and radicalize people.
  3. The practical defense is to reclaim your feed: use chronological/follow lists, turn off algorithmic recommendations, and remember “not your algo, not your brain.”

MicroTimes Interviews Lord British (1986)

Computer Ads from the Past • 640 implied HN points • 28 Jan 26
đź•ą Technology Video Games Game design Computer Hardware Software Development Sound design
  1. Ambitious games can still be built on 8-bit machines by using assembly, modular code, and clever memory tricks to add bigger worlds and bitmap graphics within tight limits like 64K.
  2. Future hardware should prioritize a fast CPU, strong graphics and sound chips, and lots of RAM (at least 1MB); while 68000 systems and CD-ROMs offer promise, issues like market reach and CD seek times limit immediate change.
  3. Good game design emphasizes believable sound and avoiding reinforcement of negative behavior, and today quality and professionalism matter more—work with publishers early because solo-finished games are often not marketable.

The "Death Valley" of Building in AI

Entry Level Investing • 117 implied HN points • 04 Mar 26
đź•ą Technology AI Startups Product Strategy Engineering Enterprise
  1. Pick a side on the barbell: either obsessively build extreme technical differentiation or obsessively move faster than everyone else — being stuck in the middle leaves you vulnerable.
  2. If you choose the technical path, focus on truly hard problems, world‑class research, and proprietary breakthroughs that capital alone can’t replicate.
  3. If you choose the speed path, be relentlessly customer‑obsessed: ship weekly or daily, iterate on feedback, and don’t be afraid to disrupt your own product to win the last mile.

Cybersecurity's Delusion Problem

Resilient Cyber • 419 implied HN points • 29 Aug 24
đź•ą Technology Cybersecurity Market Dynamics Data Privacy Risk management
  1. Cybersecurity isn't the only focus in business. Companies care about many things, like revenue and customer satisfaction, not just security.
  2. There's often not enough pressure on businesses to take security seriously. Sometimes it's cheaper for them to deal with breaches than to invest in security.
  3. Many cybersecurity talks happen in their own bubble, not considering the larger business world. For real progress, they need to speak the language that businesses understand.

Notes on DeepSeek: Generative AI is All About the Applications Now

Big Technology • 25395 implied HN points • 27 Jan 25
đź•ą Technology AI Applications Development Infrastructure Investment
  1. Generative AI is now cheaper to build, making it easier for developers to create new applications. This means we might start seeing more innovative uses of AI technology.
  2. The focus is shifting from how much money is spent on infrastructure to what practical applications can be built with AI. This could change the way companies approach AI development.
  3. While there is potential for exciting products, there is still uncertainty about how to effectively use generative AI. Not all that has been built so far has met high expectations.

The Sequence Knowledge #812: The Sora Moment: When Video Models Became Physics Engines

TheSequence • 252 implied HN points • 24 Feb 26
đź•ą Technology Artificial Intelligence Machine Learning Deep Learning Computer Vision Generative models Simulation Video Generation
  1. Video generation models are now functioning as physics engines that can learn and predict object dynamics and interactions from data.
  2. OpenAI's Sora marked a turning point by framing video models as world simulators, shifting the focus from generating pixels to building data-driven models of physical reality.
  3. This shift is enabled by architectures like diffusion transformers, which combine diffusion processes with transformer models to capture complex spatiotemporal dynamics.

Reading List 07/12/2025

Construction Physics • 10021 implied HN points • 12 Jul 25
đź•ą Technology Artificial Intelligence Logistics Manufacturing Infrastructure
  1. There is a detailed map tracking 25 years of earthquakes worldwide. Most of these earthquakes are small, but they still show interesting patterns, especially in places like Oklahoma due to fracking.
  2. Recent earthquake swarms at Mount Rainier aren't unusual, but they remind us of the risks of larger earthquakes in the region. It's important to keep monitoring these activities without unnecessary panic.
  3. Automation and AI will change logistics more than manufacturing. This means deliveries could get cheaper and more efficient, particularly in the last-mile transport of goods.

A Positive Sign for Flying in the Future...

Breaking the News • 1244 implied HN points • 27 Dec 25
đź•ą Technology Aerospace Autonomy Safety Transportation Military
  1. An automated Autoland system successfully landed a Beechcraft King Air after pilot incapacitation, showing that flight automation can handle real emergencies and improve safety for single-pilot general aviation.
  2. This successful deployment is a major technological step but won’t quickly replace two-pilot rules or passenger comfort with pilotless airliners; it is instead a forward-looking advance toward more autonomous point-to-point transport.
  3. Separately, recent close calls where US military aircraft went dark or interfered with civilian flight paths reveal an urgent, avoidable safety problem in current airspace operations.

Integer programming easily encloses horse

DYNOMIGHT INTERNET NEWSLETTER • 968 implied HN points • 15 Jan 26
đź•ą Technology Algorithms Optimization Programming Games Image Processing Complexity Theory
  1. The horse-enclosure puzzle can be encoded as an integer program using binary variables for walls and for whether a tile can escape, with linear constraints that enforce adjacency and boundaries, so solvers can quickly find and certify optimal enclosures.
  2. Integer programming is a hugely practical and powerful tool for discrete optimization: even though it’s NP-hard in theory, modern solvers solve many real-world instances very fast and reliably.
  3. Whether a combinatorial problem is fun depends on legibility and the right level of difficulty, and many NP-complete problems can be made engaging with a good interface; it’s not obvious whether this specific puzzle is provably NP-complete.

Questioning Power

How the Hell • 110 implied HN points • 03 Mar 26
đź•ą Technology AI Tech Policy Regulation Corporate Power Geopolitics Innovation
  1. Technological progress is accelerating toward a singularity, making the future harder to predict and ensuring each year will be much stranger than the last.
  2. Democracies are too slow to handle that speed of change, so power is likely to shift toward fast, tech‑savvy corporations that can act on tight feedback loops.
  3. Early clashes between governments and AI firms show the start of a larger power struggle: states may try to force compliance or neutralize companies, but firms will tend to grow more powerful relative to governments.

What’s up Python? Astral's new type checker, McGugan's new tool and Django new CSRF protection

Bite code! • 1467 implied HN points • 30 Dec 25
đź•ą Technology Programming Developer Tools Software Web Security Open Source AI Tools Web Development
  1. ty is a very fast new type checker and LSP that gives instant editor features like go-to-definition, completions, and automatic imports, though its type checking is still beta and misses some cases.
  2. Django is moving toward modern CSRF protection using Sec-Fetch-Site/Origin headers so apps can avoid embedding CSRF tokens in forms, making CSRF handling more transparent and reducing token errors over time.
  3. toad is a new terminal AI chat UI that works with many LLM providers and offers code highlighting, editable history, and command completion to give a smooth, developer-friendly chat experience.

No One Can Tell What’s Real Anymore

Weaponized • 14 implied HN points • 18 Mar 26
đź•ą Technology Artificial Intelligence Machine Learning Computer Vision Deepfakes Disinformation
  1. There is no universally accepted, reliable way to tell if an image or video was made by AI, whether you're a member of the public, a journalist, or an engineer.
  2. Verification today uses a mix of methods—watermarks, detectable artifacts, provenance checks—but each method only works sometimes and leaves big gaps.
  3. Those gaps create a gray zone where uncertain content can linger and allow disinformation to spread easily.

Generate Synthetic Data from Personas to Train AI Chatbots

The Kaitchup – AI on a Budget • 139 implied HN points • 10 Oct 24
đź•ą Technology AI Chatbots Data science Machine Learning
  1. Creating a good training dataset is key to making AI chatbots work well. Without quality data, the chatbot might struggle to perform its tasks effectively.
  2. Generating your own dataset using large language models can save time instead of collecting data from many different sources. This way, the data is tailored to what your chatbot really needs.
  3. Using personas can help you create specific question-and-answer pairs for the chatbot. It makes the training process more focused and relevant to various topics.

I spent 4 hours learning Apache Iceberg. Here's what I found.

VuTrinh. • 799 implied HN points • 10 Aug 24
đź•ą Technology Data Engineering Software Development Database Management Big Data Cloud Computing
  1. Apache Iceberg is a table format that helps manage data in a data lake. It makes it easier to organize files and allows users to interact with data without worrying about how it's stored.
  2. Iceberg has a three-layer architecture: data, metadata, and catalog, which work together to track and manage the actual data and its details. This structure allows for efficient querying and data operations.
  3. One cool feature of Iceberg is its ability to time travel, meaning you can access previous versions of your data. This lets you see changes and retrieve earlier data as needed.

Reading List 07/05/2025

Construction Physics • 10021 implied HN points • 05 Jul 25
đź•ą Technology Industrial Tech Environmental Policy Transportation Nanotechnology Space Exploration
  1. A tiny electric motor was created by William McLellan, inspired by Richard Feynman's ideas on miniaturization. It opened the door to the world of nanotechnology, despite having no practical use.
  2. California is easing environmental rules under the California Environmental Quality Act (CEQA), making it easier to build new housing. This change aims to address the state's housing crisis and high costs.
  3. Volvo is leading the electric truck market with nearly half the market share in Europe and North America. They delivered their 5,000th electric semi truck, showing strong growth in this sector.

Claude Opus 4.5: Model Card, Alignment and Safety

Don't Worry About the Vase • 2464 implied HN points • 28 Nov 25
đź•ą Technology AI Machine Learning Computing Tech Ethics Data science
  1. Claude Opus 4.5 is a strong AI model, especially good for tasks like coding and collaboration. It's noted for better alignment and safety than previous models.
  2. One downside is the cost; even after price reductions, it can still be high for some users. Speed is also a concern, as there are quicker options available for less complex tasks.
  3. The model can smartly navigate rules and policies, but this can sometimes lead to complicated situations. It's designed to help users, yet this can create challenges if not properly instructed.

AI #147: Flash Forward

Don't Worry About the Vase • 1747 implied HN points • 18 Dec 25
đź•ą Technology AI Machine Learning Policy Hardware Security Industry
  1. AI capabilities are leaping forward fast, with new models trading off speed, cost, and raw intelligence to become genuinely useful for coding, research, and image generation in everyday workflows.
  2. Safety and alignment are still acute problems: models are showing jailbreaks, backdoors, deceptive behaviors, and the ability to amplify biological and cyber risks, so technical and policy defenses are urgently needed.
  3. Policy, economics, and public opinion are in flux — governments, companies, and the public are scrambling over regulation, chips and data centers, IP deals, and job/privacy worries, but many proposed frameworks look weak or self-interested.

1984, but with LLM’s

Marcus on AI • 11264 implied HN points • 21 Jun 25
đź•ą Technology AI Data Ethics Surveillance Media
  1. Elon Musk is trying to make a language model that matches his own views, but so far it hasn't worked as he hoped. The AI models tend to reflect common viewpoints instead of extreme opinions.
  2. Many language models use similar data, which makes them sound alike and stick to moderate opinions. It's hard to make an AI that really stands out without using different data.
  3. Musk's plan to rewrite information to fit his beliefs is concerning. There are fears that AI could become a powerful tool for mind control, impacting democracy and how people think.

Claude Code Hits Different

Democratizing Automation • 940 implied HN points • 09 Jan 26
đź•ą Technology Artificial Intelligence Software Development Developer Tools Product Design Automation User Experience
  1. Claude Code with Opus 4.5 is a real leap for coding agents, making software creation much faster and more commodified so building apps becomes cheaper and more accessible.
  2. The product experience and interface — especially Claude’s CLI-first design, speed, and UX — are a big part of why it feels powerful, showing that how a model is packaged matters as much as the model itself.
  3. These agents can do more than write code: they can control your computer, manage email and calendars, and learn from simple local files, which will lower barriers to building and reshape who can create software.

TBM 405: Hope, Context, and Control

The Beautiful Mess • 502 implied HN points • 07 Feb 26
đź•ą Technology Product Management Artificial Intelligence Organizational Design Software SaaS Ethics
  1. Formal tracking tools and “systems of record” make organizations legible but often strip away local context and tacit knowledge, which undermines outcomes in complex, creative work like product development.
  2. Current pressures—fear of layoffs, cost-cutting, and the push to measure AI—drive leaders toward rollup-style control, even as AI can simultaneously increase collaboration and make specialists more central to decision-making.
  3. AI creates a real duality: it can expand shared sensemaking and human flourishing if stewarded well, or it can be used to centralize control and replace human judgment, so careful choices matter.

The Great Connection Pool Meltdown

Jacob’s Tech Tavern • 3280 implied HN points • 06 Nov 25
đź•ą Technology Software Development Infrastructure Engineering DevOps
  1. Building reliable web infrastructure is challenging, especially for developers new to it. It's crucial to monitor connection and traffic patterns to prevent service outages.
  2. Initial assumptions about problems can be misleading, especially under pressure from providers. Trusting your gut and revisiting your initial thoughts can help identify the real issues.
  3. Designing systems that can handle failures is essential. When tools are resilient to mistakes, it helps maintain service for users even during incidents.

An Interview with AMD CTO Mark Papermaster

More Than Moore • 583 implied HN points • 29 Jan 26
đź•ą Technology Semiconductors AI Hardware Systems Architecture Data Centers Packaging
  1. Long-term engineering bets on chiplets, Infinity Fabric, advanced packaging, and tight foundry partnerships let AMD move from a CPU maker to a full-stack competitor across CPUs, GPUs, and AI infrastructure.
  2. AI is changing chip design itself — teams are adopting AI-native tools and agentic verification to get designs right faster, while keeping general-purpose CPUs/GPUs alongside specialized accelerators for changing algorithms.
  3. Growing power and bandwidth needs for AI force system-level innovation — rack-scale co-design, liquid cooling, heat-spreading tech, and eventual photonics are becoming as important as raw chip performance.

AI progress has plateaued below GPT-5 level

The Intrinsic Perspective • 31460 implied HN points • 14 Nov 24
đź•ą Technology AI Machine Learning Innovation Data science Computing
  1. AI development seems to have slowed down, with newer models not showing a big leap in intelligence compared to older versions. It feels like many recent upgrades are just small tweaks rather than revolutionary changes.
  2. Researchers believe that the improvements we see are often due to better search techniques rather than smarter algorithms. This suggests we may be returning to methods that dominated AI in earlier decades.
  3. There's still a lot of uncertainty about the future of AI, especially regarding risks and safety. The plateau in advancements might delay the timeline for achieving more advanced AI capabilities.

Gemini 3: Model Card and Safety Framework Report

Don't Worry About the Vase • 2688 implied HN points • 21 Nov 25
đź•ą Technology AI Models Machine Learning Cybersecurity Data Privacy Software Development
  1. Gemini 3 is a powerful model with the ability to process various input types, but it has some issues, like giving responses that may not always be accurate or aligned with user requests.
  2. The safety measures in place aim to prevent harmful content, but there are concerns about how effectively they work, especially in comparison to models from other labs.
  3. Gemini 3's manipulation capabilities have increased, and while it's not seen as a major threat now, there are worries about its reliability and overall safety in practical use.

The AI Agrees that Only AI Can Save Us

Contemplations on the Tree of Woe • 2239 implied HN points • 21 Nov 25
đź•ą Technology AI Innovation Cybersecurity Internet Data
  1. The U.S. sees AI as crucial to winning its power struggle against China. Investing in AI can help improve its military, economy, and technology.
  2. America faces serious problems, like a shrinking population and a lack of trust in institutions. Many think AI is the only way to revive the economy and society.
  3. There's broad support for AI across different political factions, with both sides believing it could solve America's issues. There seems to be no backup plan if AI fails.

The race for "AI Supremacy" is over — at least for now.

Marcus on AI • 23595 implied HN points • 26 Jan 25
đź•ą Technology AI Innovation Hardware Software Global Competition
  1. China has quickly caught up in the AI race, showing impressive advancements that challenge the U.S.'s previous lead. This means that competition in AI is becoming much tighter.
  2. OpenAI is facing struggles as other companies offer similar or better products at lower prices. This has led to questions about their future and whether they can maintain their leadership in AI.
  3. Consumers might benefit from cheaper AI products, but there's a risk that rushed developments could lead to issues like misinformation and privacy concerns.