Best AI Twitter (X) Accounts to Follow in 2026

AI Twitter—now AI X—can alert you to a release or paper quickly. It can also turn an unverified screenshot into a consensus before anyone opens the underlying artifact.

Updated August 30, 2026 — links and affiliations verified.

The goal is not to follow the most accounts. Build a balanced feed from people and organizations that link primary artifacts, document experiments, show expertise, correct mistakes, and cover different parts of the AI ecosystem.

Definition: AI Twitter / AI X

AI Twitter or AI X is the network of researchers, engineers, founders, educators, policymakers, and official organizations discussing artificial intelligence on X. A useful account links to papers, code, documentation, evaluations, or firsthand experiments instead of merely repeating news.

TL;DR

Start with Andrej Karpathy, Simon Willison, Ethan Mollick, Sebastian Raschka, and Chip Huyen
Use official lab accounts for an announcement, then open the paper, documentation, or model card
Add research, engineering, and policy perspectives so one launch cycle does not define your feed
Treat X as discovery; verify decisions with the original artifact or a durable long-form source

For a source layer beyond social posts, use AI blogs and news sites alongside this feed.

How This List Is Maintained

Accounts are included for a clear role in a reader’s feed: primary announcements, research explanation, hands-on engineering, applied use, or policy and public-interest context. The list avoids follower counts and should be rechecked for activity, handle changes, and original-source links each quarter.

Best AI X Accounts at a Glance

This matrix includes the 30 distinct accounts recommended below. Evidence to expect is a reason to follow, not a guarantee that every post meets the standard.

Feed role	Account	Best for	Evidence to expect
Research explainer	@karpathy	LLM intuition and education	Longer explanations, talks, and experiments
Hands-on testing	@simonw	Model and API behavior	Prompts, outputs, and linked notes
Applied use	@emollick	Work and education	Research-informed practical guidance
Technical explanation	@rasbt	How language models work	Diagrams, code, and paper discussion
Engineering systems	@chipro	Production AI systems	System-design and evaluation tradeoffs
Research	@demishassabis	AI-for-science and lab research	Primary research announcements
Research	@drfeifei	Vision and human-centered AI	Research and institutional work
Research	@fchollet	Reasoning and evaluation	Benchmark and generalization analysis
Research	@hardmaru	Generative and world models	Papers and visual experiments
Research	@DrJimFan	Robotics and embodied AI	Research explanations and demos
Research	@natolambert	Open models and post-training	Research commentary and synthesis
Applied AI	@AndrewYNg	Learning and adoption	Accessible industry perspective
Engineering	@swyx	AI engineering	Tools, events, and implementation patterns
Engineering	@HamelHusain	Evaluation and LLM engineering	Practical methods and lessons
Engineering	@sh_reya	Data systems and evaluation	Reliable-AI application thinking
Engineering	@hwchase17	Agent frameworks	Primary ecosystem updates
Primary source	@OpenAI	OpenAI releases	Official announcement links
Primary source	@AnthropicAI	Anthropic releases	Official announcement links
Primary source	@GoogleDeepMind	Google DeepMind research	Research and product links
Primary source	@AIatMeta	Meta AI research	Research and model links
Primary source	@MistralAI	Mistral releases	Official model and product links
Primary source	@huggingface	Open-model ecosystem	Repository and community links
Primary source	@NVIDIAAI	AI systems and platforms	Developer and research links
Primary source	@Cohere	Enterprise AI releases	Official product and research links
Primary source	@perplexity_ai	Product releases	Official product links
Policy	@StanfordHAI	Research and policy	Institutional research links
Policy	@NIST	Standards and measurement	Official technical guidance
Policy	@OECDinnovation	International policy and data	Cross-country research links
Public interest	@AdaLovelaceInst	Governance and social impact	Public-interest research
Public interest	@AINowInstitute	Accountability and labor	Public-interest analysis

The Five Best Starting Accounts

1. Andrej Karpathy — @karpathy

Follow @karpathy. His posts often connect to longer technical explanations, repositories, talks, or experiments. Start here if you want intuition about neural networks, language models, tokenization, training, and AI-assisted coding.

2. Simon Willison — @simonw

Follow @simonw. He publishes documented notes about language-model products and APIs, including prompts and outputs. That makes the account useful when launch marketing outruns careful testing.

3. Ethan Mollick — @emollick

Follow @emollick. His work is useful for readers focused on AI in work and education rather than implementation details. Follow the linked research and longer writing when a recommendation matters to your organization.

4. Sebastian Raschka — @rasbt

Follow @rasbt. Raschka’s diagrams and code-oriented explanations help readers investigate what changed in architecture, training, post-training, or inference after a model release.

5. Chip Huyen — @chipro

Follow @chipro. Her posts are useful for evaluation, data, product architecture, latency, cost, and the difference between a demo and a dependable system.

Research, Engineering, Primary Sources, and Policy

The matrix above is the complete shortlist. Use the account groups deliberately:

Research: follow the research accounts when you need papers, evaluation context, or a technical framing that goes beyond release headlines.
Engineering: follow builder accounts when you need code, operational tradeoffs, and implementation lessons. Cross-check fast-moving framework claims with official changelogs.
Primary sources: official accounts are the quickest path to a lab’s own launch links, but they are still an organization’s perspective.
Policy and public interest: these accounts counterbalance the launch cycle with standards, regulation, social impact, accountability, and labor context.

Build a High-Signal AI X Feed

A practical starting setup is four private X Lists, not one algorithmic home feed:

Primary sources: labs, research groups, standards bodies, and official product accounts.
Research: people who publish or explain papers and evaluations.
Engineering: builders who share code, systems design, and operating tradeoffs.
Work and policy: educators, organizational researchers, and public-interest institutions.

Start small—about five to eight accounts per list—and review a list only when it serves a purpose. X is a discovery layer: open the original paper, code, documentation, evaluation, or long-form post before acting on a claim.

A Simple Quarterly Audit

Keep an account when at least two of these are true:

It regularly links primary sources.
It adds expertise you cannot get from a lab announcement.
It shows methods, prompts, code, or limitations.
It separates fact, interpretation, and prediction.
It corrects or updates earlier claims when needed.

Mute or remove accounts whose feed is mostly outrage, affiliate promotion, screenshots without context, or repeated summaries of other people’s work.

Want a weekly source-controlled reading list in addition to your X feed? Build a weekly AI article recommendation workflow that collects, deduplicates, and ranks the sources you choose.

Who is the best AI account to follow on X?

For one general-purpose follow, start with Andrej Karpathy for technical intuition or Simon Willison for documented testing of new AI products. Add Ethan Mollick when your focus is work and education rather than engineering.

Is X still useful for AI news in 2026?

Yes, especially for primary announcements and release-time discussion. Availability, reach, and post visibility can vary by account and platform settings, so use X for discovery and open the original artifact for verification.

How many AI X accounts should I follow?

Start with a small set split across primary sources, research, engineering, and work or policy. The exact number matters less than whether you can review the feed without missing evidence-bearing posts.

Should I follow AI labs or individual researchers?

Follow both. Lab accounts provide primary announcements and official links; researchers and independent builders add context, criticism, and hands-on testing. Neither group is sufficient alone.

How can I tell whether an AI influencer is reliable?

Look for primary links, disclosed methods, complete examples, clear uncertainty, and corrections. Be skeptical of accounts that publish constant superlatives, unexplained benchmarks, or affiliate-heavy tool lists.

Are follower counts a good way to rank AI accounts?

No. Follower counts measure reach, not accuracy or usefulness. A specialist who posts one careful evaluation can be more useful to your work than a large account that republishes every launch.

Bottom Line

Build a small portfolio of perspectives: primary accounts for what shipped, researchers and builders for explanation and testing, and policy institutions for wider consequences. Prune it whenever an account stops helping you inspect the evidence.