ctaio.dev Ask AI Subscribe free

AI · AI Education · AI newsletters

Best AI Newsletters in 2027

The best AI newsletters for technology leaders in 2027 are Stratechery and Exponential View for strategy, Import AI and Interconnects for research direction, and The Pragmatic Engineer, Latent Space and Simon Willison's newsletter for what is changing in engineering practice. For a daily scan, The Rundown AI and TLDR AI are the largest free options, each with more than a million subscribers by their own count. The list below has 14 newsletters grouped by the job they do: daily news, strategy and analysis, building, and research. For each one I give the audience size where the publisher states it, the cadence, the price and the published sponsorship contact, all checked on the publisher's own pages on 19 November 2026.

By · Published November 19, 2026 · Checked November 19, 2026

Printed newsletters stacked on a dark desk under warm amber lamp light

Key takeaways

  1. Read one daily, one strategy title, one builder title and one research digest. More than four newsletters becomes an unread folder.
  2. Pay for at most two. Stratechery and The Pragmatic Engineer cost $150 a year each and are the easiest to justify on a leadership budget.
  3. Audience numbers are self-reported by publishers, often on advertiser pages. Use them to judge reach, not quality.
  4. Check each newsletter's sponsorship model. The Pragmatic Engineer and Interconnects run no ads, while the big dailies are ad-funded and carry sponsor slots in every issue.

Which AI newsletters are worth a technology leader's time?

Ranked within each group by how much a technology leader gets out of it. "Checked" says where each number came from and when.

Daily news

  1. 2.1M subscribers (publisher) · daily · free

    For: Leaders who want one five-minute daily scan of launches, funding and product news.

    The largest AI news daily by its own count, written for a broad business audience rather than engineers. It is fast and well edited, and it tells you what shipped yesterday; it rarely tells you whether it matters to your stack. Pair it with one slower analysis title.

    Checked: Publisher advertise page (2.1M AI newsletter subscribers, 3.5M+ readers across titles), 19 Nov 2026

  2. 1.1M+ subscribers (publisher) · every weekday · free

    For: Engineering managers and staff engineers who want links to papers, repos and launches, not commentary.

    Short summaries with a link out for every item, which makes it the most skimmable of the dailies and the one most likely to surface a GitHub repo or paper your team will ask about. Coverage leans technical. Expect a sponsor slot in every issue.

    Checked: Publisher site and advertise page (homepage says 1.2M+, advertise page lists about 1.15M for TLDR AI), 19 Nov 2026

  3. 1M+ subscribers (publisher) · daily · free

    For: Executives outside engineering, and leaders who want to forward practical AI usage tips to their teams.

    Written by Zain Kahn and team, it mixes news with prompts and workflow tips for knowledge workers. Its advertise page says more than 200,000 US subscribers are C-level, VP or Director, which makes it a useful read on what your business peers are seeing. Light on engineering depth.

    Checked: Publisher site and advertise page (homepage copy says 2M+, advertise page lists 1M+ for this title), 19 Nov 2026

Strategy and analysis

  1. Subscriber count not published · free weekly article, 3 paid updates a week · $15/month or $150/year

    For: CTOs, CIOs and CAIOs who brief boards on platform shifts, pricing power and vendor strategy.

    Ben Thompson's business-model analysis of the big platforms is the closest thing to a shared vocabulary among technology executives, and most of it is now about AI. The paid Stratechery Update is where the timely analysis lives. Since October 2026 the site is presented by Stripe, with an editorial independence clause disclosed on the About page.

    Checked: Publisher About and Stratechery Plus pages, 19 Nov 2026

  2. 168,000+ readers (publisher) · weekly · free, paid $19/month or $120/year

    For: CAIOs, strategy leads and anyone who has to connect AI to energy, capital markets and policy.

    Azeem Azhar writes about AI as an economic and industrial shift, with data on compute, energy and adoption that travels well into a board deck. Paid members get a Sunday briefing, a Monday data read and the essays. Less useful for day-to-day engineering decisions.

    Checked: Publisher About page, 19 Nov 2026

  3. Hundreds of thousands of subscribers (publisher wording) · a post every two to three weeks recently · no price listed

    For: Executives leading AI adoption across non-technical functions.

    Wharton professor Ethan Mollick writes about how AI changes knowledge work, with hands-on comparisons of current models and frequent guides on which model to use for what. It is the newsletter I would forward to a CEO or CHRO. The About page says he writes every post himself.

    Checked: Publisher About page and archive (post dates May to Nov 2026), 19 Nov 2026

  4. Hundreds of thousands of subscribers (publisher wording) · weekly · free, optional paid

    For: CAIOs and research leads who need early warning on capability and policy shifts.

    Jack Clark, a co-founder of Anthropic, summarises research papers each week and adds his view on what they imply for capability, safety and policy. It is the best way I know to see which papers frontier labs are paying attention to. Read it knowing the author's affiliation.

    Checked: Publisher About page on Substack, 19 Nov 2026

  5. 180,000+ subscribers (publisher) · weekly · free and paid tiers

    For: CTOs and infrastructure leaders making GPU, neocloud and data centre commitments.

    Dylan Patel's team covers chips, supply chains and AI data centres in more depth than any other newsletter, with cost models behind the claims. The free posts are worth reading; the paid tier gives access to every article, and the institutional research is sold separately. Overkill if you only rent capacity from a hyperscaler.

    Checked: Publisher About page and FAQ (180,000+ subscribers, group discount 20%); paid price not listed on those pages, 19 Nov 2026

For builders

  1. Subscriber count not published · AINews every weekday, main newsletter 2 to 3 times a week · free, paid Latent Space Pro

    For: VPs of Engineering, platform leads and AI engineers building agents and LLM features.

    The publication of the AI engineer community, started by swyx, covering agents, evals, inference and coding tools with interviews from the people building them. AINews, the weekday digest that used to live at news.smol.ai, is now part of Latent Space, and the full email version sits in the paid tier.

    Checked: Publisher About and news pages (news.smol.ai now redirects to latent.space/news), 19 Nov 2026

  2. 1M+ readers (publisher) · deep dive Tuesdays, paid The Pulse Thursdays · free, paid $15/month or $150/year

    For: CTOs and engineering leaders deciding how AI coding tools change teams, hiring and process.

    Gergely Orosz reports on how engineering organisations actually work, and a growing share of that is about AI tooling and its effect on teams. The newsletter carries no sponsorships or ads, which matters when the subject is tool choice. The paid tier is the one I would expense first for an engineering leadership team.

    Checked: Publisher About page, 19 Nov 2026

  3. Subscriber count not published · about weekly · free, paid monthly briefing from $10/month

    For: Hands-on technical leaders who still want to try the tools themselves.

    Co-creator of Django and creator of Datasette, Simon Willison tests new models and tools in public and writes up what works, what breaks and where the security holes are, especially prompt injection. The free newsletter is a digest of his blog. Since February 2026 he sells one sponsor slot a week and states sponsors have no editorial influence.

    Checked: Author's blog About page and sponsorship post (Feb 2026), 19 Nov 2026

Research

  1. Subscriber count not published · weekly · free

    For: Leaders who want a calm weekly research and industry roundup with an expert framing.

    DeepLearning.AI's weekly newsletter opens with a personal letter from Andrew Ng and then explains a handful of research and industry stories in plain language. It is slower than the dailies and better for understanding why a result matters. A good default for managers moving into AI roles.

    Checked: Publisher The Batch pages (weekly, letter from Andrew Ng); free per DataCamp listing (third-party), 19 Nov 2026

  2. 200,000+ readers (publisher) · long-form issues, no fixed schedule stated

    For: Technical leaders who want to understand how LLM architectures and training methods are changing.

    Sebastian Raschka, author of Build a Large Language Model (From Scratch), writes detailed explainers on new model architectures, reasoning methods and research papers. It is the most technical title here and the most durable: issues stay useful for months. Best for a staff engineer or head of AI rather than a CIO.

    Checked: Publisher About page and author contact page, 19 Nov 2026

  3. Tens of thousands of subscribers (publisher wording) · 1 to 3 times a week · free, paid tier with group discounts

    For: Heads of AI and research leads weighing open models, post-training and model selection.

    Nathan Lambert, who led post-training for the Olmo models at Ai2, writes model reviews, open-model surveys and essays from a practitioner's position. It is the strongest single source on open-weight models. The About page says it takes no advertising.

    Checked: Publisher About page; paid price not listed there, 19 Nov 2026

How should a technology leader build an AI reading stack?

Build it like a portfolio with four slots, each doing a different job. A daily title tells you what shipped, a strategy title tells you what it means for your market, a builder title tells you what it changes in your team's work, and a research digest tells you what is coming in six to eighteen months. The common mistake is three dailies and nothing in the other slots.

For the daily slot, pick one of The Rundown AI, TLDR AI or Superhuman AI and unsubscribe from the others. They cover largely the same launches, and reading two of them doubles the time without doubling the signal. TLDR AI suits engineering leaders because it links out to repos and papers; The Rundown AI and Superhuman AI suit leaders who need business framing they can forward.

The other three slots are where the value sits. A weekly hour with Stratechery or Exponential View, Latent Space or The Pragmatic Engineer, and Import AI or The Batch gives you more decision-ready context than a month of daily skims. If you only have time for one newsletter in total, take the builder or strategy title and drop the daily, because the daily news will reach you anyway through your team and your feeds.

Newsletters are a filter, and the primary document still decides. When a decision depends on a model's performance, read the model card, the eval report or the vendor's pricing page yourself, and be careful with leaderboard claims quoted secondhand (see benchmaxxing). For the same reason, pair your reading with the people worth following on X and the YouTube channels that show tools in action.

Which AI newsletters fit which leadership role?

The right mix depends on what you are accountable for. A CTO or VP of Engineering running teams that write code with agents should anchor on The Pragmatic Engineer, Latent Space and Simon Willison: between them they cover organisational effects, the AI engineering community and hands-on tool testing, including the security problems that matter for agentic engineering in production.

A CAIO or CIO who owns the AI portfolio and the board conversation gets more from Stratechery, Exponential View and One Useful Thing. Stratechery explains platform and pricing moves, Exponential View connects AI to energy, capital and policy, and Ethan Mollick's posts are the clearest writing available on how AI changes non-technical work. The Chief AI Officer guide covers what else that role has to track.

Leaders who sign infrastructure contracts should add SemiAnalysis. Its coverage of chips, power and data centre build-outs is the most detailed in newsletter form, and it is the right companion to our pages on neoclouds and compute capacity planning. If you only consume cloud APIs, the free posts are enough.

Heads of AI and research leads should read Import AI, Interconnects and Ahead of AI. Import AI shows which papers frontier labs care about, Interconnects is the best source on open-weight models and post-training, and Sebastian Raschka's explainers are the ones to hand an AI engineer who needs to understand a new architecture properly.

Are paid AI newsletters worth the money?

A paid newsletter is worth it when it changes one decision a year, and for the best titles that bar is low. Stratechery Plus costs $15 a month or $150 a year, The Pragmatic Engineer the same, and Exponential View $19 a month or $120 a year. Against a single vendor negotiation or hiring decision, none of these prices registers.

The case is strongest for team subscriptions. Interconnects offers group discounts of 10% to 25% by seat count, SemiAnalysis offers 20% off for groups, and Exponential View provides an email template for asking a manager to expense it. Buying a builder title for your engineering managers creates a shared reference point, which is worth more than one leader reading alone.

Where I would not pay: a second daily, or a premium tier of a news digest. Daily news loses value within days, and the free tiers of The Rundown AI, TLDR AI and Superhuman AI already carry the headlines. Spend the budget on analysis that stays relevant for a quarter, and cancel any subscription you have not opened in a month.

How should you read sponsored AI newsletters?

Most large AI newsletters are funded by advertising, and the advertiser pages show who the sponsors are trying to reach. Superhuman AI says more than 200,000 of its US subscribers are C-level, VP or Director; The Rundown AI and TLDR AI publish reach and open-rate figures for buyers. That is useful context for you: if a tool appears in three dailies in one week, it may be a media buy rather than a groundswell.

Disclosure practices differ and are worth checking before you trust a recommendation. The Pragmatic Engineer runs no ads or sponsorships in the newsletter, and Interconnects says it has accepted no advertising money. Stratechery has been presented by Stripe since October 2026 and discloses that in relevant posts, and Simon Willison has sold a weekly sponsor slot since February 2026 with a public statement that sponsors have no editorial influence.

The practical rule: treat tool roundups and "top tools" sections in ad-funded dailies as leads to investigate, never as evaluations. For anything that touches procurement, run your own trial and use a structured process like the one in our guide to buying AI.

How were these newsletters chosen and checked?

I selected for usefulness to people who run technology organisations: each title has to either sharpen a strategic judgement or change how a team builds. Popularity alone did not qualify a newsletter, and several large consumer-oriented AI digests were left out because they overlap with the three dailies listed.

Every audience figure, cadence and price comes from the publisher's own site or advertiser page, checked on 19 November 2026, and where a publisher gives only a phrase such as "hundreds of thousands" I quote the phrase rather than invent a number. Where a homepage and an advertiser page disagree, I use the more conservative figure and say so in the item. Contacts are the sponsorship or business addresses each publisher lists; where none is published, the field links to the relevant page or stays empty.

Listed here, or should be?

If your entry is out of date or you think something belongs on this list, email [email protected]. Entries are editorial and free. A featured, labelled Sponsored slot is available through vendor listings.

Related questions

More from AI Education

Frequently Asked Questions

What is the best AI newsletter for a CTO?
For most CTOs, The Pragmatic Engineer is the best single choice because it covers how AI tools change engineering teams and runs no ads. Add Stratechery if you also brief the board on platform and vendor strategy, and Latent Space if your teams build agents or LLM features.
Which free AI newsletter is best for daily news?
TLDR AI is the best free daily for engineering leaders because each item links to the underlying repo, paper or announcement. The Rundown AI is the largest by subscribers (2.1M by its own count) and suits leaders who want business framing. Pick one; they overlap heavily.
How many AI newsletters should an executive subscribe to?
Four is a sensible ceiling: one daily, one strategy title, one builder title and one research digest. Beyond that, issues go unread and the newsletters stop working as a filter.
Are AI newsletter subscriber numbers reliable?
They are self-reported, often on pages written for advertisers, and they are not audited. Treat them as a measure of reach. Several publishers on this list give only a phrase such as "hundreds of thousands", which is what this page quotes.
Which AI newsletter covers AI infrastructure and chips?
SemiAnalysis is the most detailed newsletter on chips, supply chains and AI data centres, with more than 180,000 subscribers by its own count. Exponential View covers the energy and capital side at a more strategic level.
·
Thomas Prommer
Thomas Prommer Technology Executive — CTO/CIO/CTAIO

These salary reports are built on firsthand hiring experience across 20+ years of engineering leadership (adidas, $9B platform, 500+ engineers) and a proprietary network of 200+ executive recruiters and headhunters who share placement data with us directly. As a top-1% expert on institutional investor networks, I've conducted 200+ technical due diligence consultations for PE/VC firms including Blackstone, Bain Capital, and Berenberg — work that requires current, accurate compensation benchmarks across every seniority level. Our team cross-references recruiter data with BLS statistics, job board salary disclosures, and executive compensation surveys to produce ranges you can actually negotiate with.