The hottest AI Substack posts right now

And their main takeaways
Category
Top Technology Topics

Links for June

Bullfish Hole • 58 implied HN points • 03 Jun 23
🕹 Technology AI
  1. AI technology can be used for both dishonest and creative purposes.
  2. Economics attracts serious individuals, and empirical work in the field involves less p-hacking compared to other disciplines.
  3. Child mortality rates have significantly declined over time, reflecting societal changes and advances in healthcare.

Prompt Engineering Guide (Web Version)

Elvis's Blog • 58 implied HN points • 17 Mar 23
🕹 Technology AI
  1. New web version of Prompt Engineering Guide launched with lectures, notebooks, and latest AI papers.
  2. Added section on models like GPT-4 and ChatGPT to showcase capabilities and limitations.
  3. Includes notebooks for starting with prompt engineering using tools like openai and LangChain.

Synthetic Insights With GPT-4

Addition • 58 implied HN points • 05 Apr 23
🕹 Technology AI
  1. Use high-quality data to ground AI in generating insights.
  2. Show AI examples of the insights you want it to generate.
  3. Scale the process by generating many insights and identifying the best ones.

🔮 Rise of Exponential Orgs, 🔑 Candidate Experiences, 🌎 Global Attitudes on AI, 🏔️ GenZ Expectations vs Reality

Work3 - The Future of Work • 58 implied HN points • 20 Jun 23
💼 Business AI
  1. Organizations are adapting to AI by becoming more efficient and leveraging new technologies.
  2. Improving candidate experiences is crucial, as engagement levels are high but attraction and conversion could be better.
  3. Accenture's $3 billion investment in AI shows the increasing importance of AI in the future of work.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

AI anthropomorphism; Marshall Brain's "Manna"; further reading

Paola Writes • 58 implied HN points • 21 May 23
🕹 Technology AI
  1. Researchers are discussing the risks of AI anthropomorphism and the importance for designers and developers to take responsibility.
  2. The novel 'Manna' by Marshall Brain explores two contrasting views of humanity's future with AI.
  3. Various perspectives from writers and researchers highlight the intersection of AI with society and capitalism, urging policymakers to take action.

Protecting Patient Privacy with AnonCAT

AI for Healthcare • 58 implied HN points • 26 Apr 23
🏥 Health & Wellness AI
  1. Protecting patient privacy involves removing or masking Personal Health Information (PHI)
  2. AI models should not learn from identifiable data to ensure patient privacy
  3. Deep learning models like AnonCAT offer an adaptable solution for accurately redacting Electronic Health Records

Recommendations for March 28, 2023

Extropic Thoughts • 58 implied HN points • 29 Mar 23
🔬 Science AI
  1. Water found on the moon in tiny glass beads may help future space missions.
  2. New propulsion concept by NASA could reach interstellar space in under 5 years.
  3. Discussions on nuclear power covering safety concerns, economics, and weapons proliferation.

Will AGI Emerge from Large Language Models?

Yuxi’s Substack • 58 implied HN points • 28 Feb 23
🕹 Technology AI
  1. AGI, or Artificial General Intelligence, is a major goal in the field of AI.
  2. Language models like GPT-3 have shown impressive abilities but still lack full functional competence.
  3. Approaching AGI through large language models may involve integrating language processing with perception, reasoning, and planning.

The Sequence Research #490: A Practical Deep Dive Inside DeepSeek-R1

TheSequence • 70 implied HN points • 14 Feb 25
🕹 Technology AI
  1. DeepSeek-R1 is a new AI model that performs well without needing to be very big. It uses smart training methods to achieve great results at a lower cost.
  2. The model successfully matches the performance of a larger, more expensive model called GPT-o1. This shows that size isn't the only thing that matters for good performance.
  3. DeepSeek-R1 challenges the idea that you always need large models for reasoning, suggesting that clever techniques can also lead to impressive results.

The Most Valuable Problem in AI

How the Hell • 313 implied HN points • 30 Aug 23
🕹 Technology AI
  1. In AI, there's a shift to being able to throw any amount of compute power at problems
  2. We are approaching a world where we can solve any intellectual problem by allocating money as a compute budget to AI agents
  3. Solving the problem of efficient compute allocation can lead to building the most valuable company of the century

Edge 439: SSMs with Attention, Understanding Zamba

TheSequence • 112 implied HN points • 15 Oct 24
🕹 Technology AI
  1. Combining state space models (SSMs) with attention layers can create better hybrid architectures. This fusion allows for improved learning capabilities and efficiency.
  2. Zamba is an innovative model that enhances learning by using a mix of Mamba blocks and a shared attention layer. This approach helps it manage long-range dependencies more effectively.
  3. The new architecture reduces the computational load during training and inference compared to traditional transformers, making it more efficient for AI tasks.

Of Agents and Agency

Gradient Ascendant • 7 implied HN points • 30 Nov 25
🕹 Technology AI
  1. LLMs and agents produce helpful outputs, but those outputs are tools — first drafts or prototypes — that almost always need verification and editing before they become real solutions.
  2. Real agency comes from expertise, and AI won’t give you that for free; treating AI outputs as finished products often creates the illusion of agency and leads to mistakes.
  3. For people with expertise, AI agents are powerful force multipliers, and although future planning agents might coordinate sub-agents more reliably, for now AI mainly accelerates expert work rather than replacing it.

The Sequence Engineering #474: The Super Popular Eliza Framework for Building AI Agents

TheSequence • 77 implied HN points • 22 Jan 25
🕹 Technology AI
  1. The Eliza framework is becoming very popular, especially in the web3 and crypto spaces. It helps developers create AI applications by automating essential tasks.
  2. Despite not being widely known, Eliza has gained a lot of attention on platforms like GitHub, showing its growing appeal.
  3. Eliza offers a flexible design, making it a strong choice for building agentic apps. It's more than just a tool for crypto; it's useful for various types of AI projects.

Workplace AI just had a terrible week

Make Work Better • 92 implied HN points • 26 Nov 24
🕹 Technology AI
  1. Microsoft's Copilot AI has faced serious criticism recently, with many users finding it unreliable and disconnected from actual business needs. Less than 4% of IT leaders reported that it provided significant value, raising concerns about its effectiveness.
  2. There are issues with Copilot accidentally accessing and sharing sensitive company information. This has created trust problems, as employees worry about privacy and data security.
  3. Next year, companies are moving towards 'agentic AI', where AI not only assists but takes on tasks autonomously. This shift aims to improve efficiency, but it's crucial to ensure these systems remain secure and trustworthy.

The Coming Wave of AI, and How Nvidia Dominants

Mule’s Musings • 366 implied HN points • 30 May 23
🕹 Technology AI
  1. Large Language Models (LLMs) are powering AI applications and depend on factors like model size, training data, and computing power.
  2. Semiconductors benefit from the demand for LLMs due to their computing power requirements for training and inference, creating opportunities for companies like Nvidia.
  3. Nvidia dominates in the AI hardware market with a three-headed hydra strategy focusing on networking and systems, accelerator hardware, and software solutions.

A Tale of a Close Partnership

Sector 6 | The Newsletter of AIM • 39 implied HN points • 20 Dec 23
🕹 Technology AI
  1. AMD has partnered with Lamini to help startups create and run generative AI products using AMD GPUs. This collaboration started in September and aims to address the GPU shortage in the AI industry.
  2. Lamini disclosed that they have been exclusively using AMD GPUs for the past year, showcasing their commitment to this partnership. They even highlighted their continuous use of AMD hardware at an AI event.
  3. Together, AMD and Lamini have developed the LLM Superstation, a powerful supercomputer equipped with 128 AMD Instinct GPUs. This setup allows businesses to train large AI models more effectively.

And the antidote to polarization is... deliberative tech

techandsocialcohesion • 19 implied HN points • 26 Mar 24
🕹 Technology AI
  1. Deliberative technology, enhanced by AI, can foster inclusive public discourse by bringing together diverse perspectives to tackle shared challenges.
  2. Deliberative technologies enable dynamic exchanges that go beyond traditional polls, allowing participants to refine solutions collaboratively.
  3. The integration of AI in deliberative tech not only streamlines processes but also amplifies democratic participation, navigates polarization, and reveals common ground for more effective solutions.

How to build an AI agent that turns your blog posts into Linkedin posts

Market Curve • 28 implied HN points • 23 Jul 25
🕹 Technology AI
  1. You can use an AI agent to automatically turn your blog posts into LinkedIn posts in just a few seconds. This saves you time and helps you share content without extra effort.
  2. To set up this system, you need to connect tools like n8n and Firecrawl to scrape your blog content and then send it to an AI to create LinkedIn posts.
  3. The process is designed for beginners, so you won’t need any coding skills. Just follow the simple steps to create a working workflow.

DRAGIN: Dynamic RAG Based On Real-Time Information Needs Of LLMs

Cobus Greyling on LLMs, NLU, NLP, chatbots & voicebots • 19 implied HN points • 26 Mar 24
🕹 Technology AI
  1. Dynamic Retrieval Augmented Generation (RAG) improves the way information is retrieved and used in large language models during text generation. It focuses on knowing exactly when and what to look up.
  2. Traditional RAG methods often use fixed rules and may only look at the most recent parts of a conversation. This can lead to missed information and unnecessary searches.
  3. The new framework called DRAGIN aims to make data retrieval smarter and faster without needing further training of the language models, making it easy to use.

The Future of Tech: Less-Hyped AI & Vibe Coding Out of Critical Systems?

ppdispatch • 8 implied HN points • 25 Nov 25
🕹 Technology AI
  1. Linus Torvalds thinks vibe coding can be useful for learning but shouldn't be used for important software projects. It's a fun way for beginners to experiment, but it can lead to maintenance problems later.
  2. Cloudflare experienced a major outage that affected many popular services like X and OpenAI due to a faulty feature file. This highlights the fragility of web infrastructure and the need for robust systems.
  3. Google is tightening security for Android developers due to rising scams. They're making it easier for students and hobbyists to experiment while also ensuring that bad actors can't easily distribute harmful apps.

Taxonomy of Red Teams

Locks and Leaks • 39 implied HN points • 19 Dec 23
🕹 Technology AI
  1. Red Teams exist to test and improve important systems, often related to cybersecurity, physical security, and decision-making.
  2. Red Teaming can be categorized into Critical Systems Testing (CST) and Applied Critical Thinking (ACT), with multiple types of red teams within each category.
  3. Collaboration among red teams is crucial, with various ways to work together such as conducting joint trainings, attending conferences, and sharing knowledge.

Column level lineage is out: AI is in

The Orchestra Data Leadership Newsletter • 39 implied HN points • 19 Dec 23
🕹 Technology AI
  1. Column-level lineage tools were popular in 2021 but might be replaced by AI for debugging data pipelines more efficiently.
  2. AI models like GPT can quickly pinpoint reasons for test failures and offer actionable insights beyond what traditional lineage tools provide.
  3. Services integrating AI with metadata can give better visibility and accurate debugging solutions for data and analytics engineers compared to column-level lineage tools.

Addressing the Hathi in the Room

Sector 6 | The Newsletter of AIM • 39 implied HN points • 19 Dec 23
🕹 Technology AI
  1. Sarvam.ai recently launched OpenHathi, a new Hindi language model that surprised many in the tech industry.
  2. They raised $41 million in funding to develop language models for various Indian languages.
  3. OpenHathi uses Meta's Llama 2 model and plans to add support for nine to ten more Indic languages soon.

High-cost AI is potentially bullish

Alex's Personal Blog • 32 implied HN points • 03 Jul 25
💼 Business AI
  1. High-cost AI tools like Perplexity and OpenAI are now charging much more for premium features, signaling a shift in how AI services are valued. As companies raise prices, it suggests they believe they offer significant value to users.
  2. Despite adding jobs in June, many industries still showed little growth, reflecting potential weaknesses in the overall labor market. This static situation raises questions about how strong employment will be in the future.
  3. Companies like Microsoft are laying off staff despite strong profits, indicating a change in how businesses view hiring. This trend could mean employers prioritize productivity from fewer employees, which may change the job market dynamics.

India's LLM Moment is Here

Sector 6 | The Newsletter of AIM • 39 implied HN points • 18 Dec 23
🕹 Technology AI
  1. Indian companies are launching new large language models (LLMs) like BharatGPT and OpenHathi, showcasing exciting developments in AI.
  2. Ola's Krutrim is unique because it's not just using existing models but creating its own LLMs and the technology to support them from scratch.
  3. These advancements in AI technology could have a big impact on various sectors, highlighting India's growing role in the global AI landscape.

Weekly Dose of Optimism #76

Not Boring by Packy McCormick • 229 implied HN points • 12 Jan 24
🕹 Technology AI
  1. Figure demonstrated a robot learning to make coffee based on observing humans, showcasing a general purpose AI approach.
  2. Rabbit introduced an AI model that translates human directives into actions without traditional app interfaces, paired with affordable hardware.
  3. US saw a decline in greenhouse gas emissions in 2023 despite economic growth, emphasizing the need for continued emissions reduction efforts.

💾 E14: The Real AI Bottleneck: High Bandwidth Memory (HBM) 💾

Let Us Face the Future • 158 implied HN points • 08 Sep 23
🕹 Technology AI
  1. High Bandwidth Memory (HBM) is crucial for datacenter AI accelerators and large language models due to its high bandwidth, low latency, and low power consumption.
  2. HBM is commercially viable, but cost and complexity remain restraints, making it more suitable for high-performance computing and AI rather than mainstream applications.
  3. The future growth of HBM depends on reducing costs, advancing technology like through-silicon vias, and addressing challenges like thermal management for wider adoption beyond datacenter and HPC.

Edge 437: Inside BlackMamba, One of the Most Important SSM Models Ever Created

TheSequence • 112 implied HN points • 08 Oct 24
🕹 Technology AI
  1. BlackMamba combines two powerful AI techniques: mixture-of-experts (MoEs) and state space models (SSMs). This helps it process long sequences and solve various AI tasks more effectively.
  2. The Mamba SSM is known for its efficiency, and BlackMamba builds on that strength while improving performance with MoE strategies.
  3. The creator is starting a new company focused on AI evaluation and benchmarking, looking for team members with expertise in these areas.

Stop begging for JSON

Artificial Ignorance • 88 implied HN points • 12 Dec 24
🕹 Technology AI
  1. Using AI tools has gotten better with structured outputs, which ensures that AI responses follow a specific format. This means developers can rely more on AI results.
  2. OpenAI introduced features like JSON mode and Structured Outputs, making it easier for developers to get the correct data structure from the AI. This reduces errors and makes integration smoother.
  3. Even with improvements, some challenges like inconsistent names and types in data still exist. Developers need to be aware and manage these issues when using AI.

Edge 454: Meet Magentic-One, Microsoft's New Framework for Building Multi Agent Systems

TheSequence • 91 implied HN points • 05 Dec 24
🕹 Technology AI
  1. Microsoft has introduced a new framework called Magentic-One for building multi-agent systems. It allows different AI agents to work together on tasks that can change or evolve.
  2. This framework is built upon another Microsoft technology called AutoGen, which helps agents collaborate effectively. It aims to manage tasks using information from the web and files from various fields.
  3. Magentic-One is part of a growing trend in AI where multi-agent systems are gaining popularity. This reflects the diverse and innovative landscape of AI development today.

Internet Independence Day

Alex's Personal Blog • 32 implied HN points • 01 Jul 25
🕹 Technology AI
  1. Cloudflare's new 'Pay Per Crawl' system lets websites charge AI bots to access their content. This could change how the internet works, making it easier for sites to earn money from their information.
  2. California's pension fund, CalPERS, is facing issues due to poor-performing investments in venture capital. Many retired workers are concerned that their pensions may be affected by these losses.
  3. Salesforce is becoming more efficient by using AI tools, which could significantly improve productivity. This efficiency may help the company grow revenues without needing as many workers.

Google Developer Knowledge MCP, Linq $20M, FireCompass Explorer

The API Changelog • 1 implied HN point • 10 Feb 26
🕹 Technology AI
  1. APIs are becoming the primary interface for AI and autonomous agents, shifting design and product decisions away from human‑first experiences. This lets assistants live inside existing apps and enables real‑time capabilities like voice translation.
  2. As APIs power more automation, security risks and supply‑chain exposure grow—hidden endpoints and misconfigurations can leak credentials across systems. Teams need proactive, agentic testing and stronger access controls to find and fix shadow APIs before attackers do.
  3. Enterprises are packaging complex domains behind unified APIs and tools to make AI integration practical across industries. Measuring AI‑readiness and centralizing documentation and access is becoming essential for reliable, maintainable integrations.

What I'm preparing for in 2025

next big thing • 76 implied HN points • 08 Jan 25
🕹 Technology AI
  1. AI is becoming a big part of software development, allowing small teams to create successful products quickly and efficiently. By 2025, we will see a lot more companies thriving because of this.
  2. We are moving towards using AI not just as helpers but as real team members. In 2025, AI will be more about collaboration rather than just assistance.
  3. There will be breakthroughs in other technologies like healthcare or energy that could surprise us, just as AI did in the past. These advancements will create new opportunities for startups.

The Alliance for the Future Manifesto

From the New World • 199 implied HN points • 12 Mar 24
🕹 Technology AI
  1. The Alliance for the Future opposes blind panic and over-regulation around artificial intelligence, aiming to educate and advocate for the benefits of AI in society and politics.
  2. AI is a process, not an object, and regulating it is complex and infeasible. History shows that negative actions should be condemned, not the technology itself.
  3. Encouraging open source development in AI can lead to a diverse range of models, efficient training, and easier detection and prevention of issues, benefitting all involved.