Each week, I organize approximately 400 AI headlines by category. Each category has a landing page. When you visit these category pages, note that the links and descriptions are pulled directly from tweets or articles, so it’s not my voice. It’s raw information, because I don’t have time to polish beyond organizing. I put all of my effort into the weekly recap AI News of the Week, which is hand-written and meant to be a readable summary of everything important for that week, with videos and images and links.
However, if you are passionate about a topic and want to follow one particular element of AI (i.e. robotic embodiment or locally hosted models), you can bookmark the category URL: https://ethanbholland.com/category/ai/robotics-embodiment/
Here is a quick walk-through of the categories, with a layperson’s descriptions:
Agency/Agents/Copilots News of the Week: Agency is when AI can do things for you (like Googling an actress name or fetching the latest weather forecast). An agent is one step further, when AI given autonomy to take action on your behalf (“Alexa, book a reservation for three at Peak in Hudson Yards for Friday night”). A co-pilot is an assistant (like spell check or autofill).
This week’s latest agent news: https://ethanbholland.com/category/ai/agents-and-copilots/
AI Inn of Court News of the Week: This category is my legal-focused lane: how lawyers, judges, and law students are actually using (and wrestling with) AI. I’m proud to serve on the executive committee of the AI Inn of Court. Expect tools like legal research assistants, document review, citation checking, discovery workflows, and the ethics questions that come with all of it. This week’s Inn of Court news: https://ethanbholland.com/category/ai/ai-inn-of-court/
Alibaba News of the Week: This category tracks Alibaba’s AI work, especially the Qwen model family. If you want to follow the “frontier but not always in the U.S. headlines” side of open models, multilingual models, and fast-moving releases, this is a good one. You’ll also see overlap with Open Source, Multimodality, Audio, and Agents… Qwen touches a lot of categories. This week’s latest Alibaba news: https://ethanbholland.com/category/ai/alibaba/
Alignment News of the Week: Alignment is the messy, important question of “how do we get AI to behave the way we want?” This includes safety training, guardrails, refusal behavior, model values, evaluation of harmful outputs, and the tradeoffs between being helpful, being honest, and being safe. If you want a fantastic introduction to alignment, I recommend this interview with Amanda Askill from Anthropic. This category often pairs with Ethics and Security, but I keep Alignment separate because it’s increasingly its own fast-moving field.
This week’s latest alignment news: https://ethanbholland.com/category/ai/alignment/
Amazon News of The Week: Most Amazon stories end up in the category they naturally belong to (agents, images, audio, robots, etc.). But when Amazon drops something big, especially on the AWS side, data centers, or proprietary models, I’ll file it here so it’s easy to follow the company’s overall AI direction.
This week’s latest Amazon AI news: https://ethanbholland.com/category/ai/amazon/
Anthropic News of the Week:
Anthropic is a top frontier AI company that builds LLMs like OpenAI, Mistral, Meta, etc. Their main AI brand is Claude. They are a critical player in shaping AI’s future, especially re regulation and ethics. As with Amazon and Apple, individual Anthropic company posts will often be placed in the categories they match (image, audio, agents, robots, etc). Anthropic focuses on alignment, safety, and transparency. CEO Dario Amodei’s interview with Lex Fridman is a good intro as is his essay Machines of Loving Grace.
This week’s Anthropic news: https://ethanbholland.com/category/ai/anthropic/
Apple News of the Week: Individual Apple products will often be filed under the categories they match, like Images, Audio, Agents, and Robotics.
Apple is currently infamous for apparently missing the boat on large language model developments. However, nearly two years ago, Apple published papers describing internal models called Large Action Models… systems designed to let iOS interact with apps without opening them. With language commands, an action model could manipulate data inside apps in the background, making the user interface optional.
That would be a huge step toward tool use and agency for Siri. It would also disrupt the app ecosystem: fewer app opens could mean fewer page views and fewer ad impressions, which would challenge any business model built on in-app advertising. I wrote an article about this called “Apple is Pulling a Braveheart” if you want a deep dive. I still think Apple has a big trick up its sleeve and is waiting for the right moment.
This weeks’ latest Apple AI news: https://ethanbholland.com/category/ai/apple/
Artificial General Intelligence (AGI) News of the Week: Artificial General Intelligence, in a nutshell, is when artificial intelligence is able to beat humans at everything (including embodying physical forms and completing physical tasks). It’s usually a thought catalyst for predictions, like when AGI will occur. 10 years? 25 years? 100? AGI is an event horizon that is tough to define, tough to imagine, and tough to predict. OpenAI defined AGI in its charter as “highly autonomous systems that outperform humans at most economically valuable work”. OpenAI has a section of its website dedicated to AGI. Google’s DeepMind published my favorite report on the five levels of artificial intelligence on the way to AGI (see also here).
This week’s latest Artificial General Intelligence (AGI) news: https://ethanbholland.com/category/ai/agi/
AI Audio News of the Week: In this case, AI audio can mean a few things. The first is “generative audio” which refers to creating sounds with AI, much like ChatGPT writes words or MidJourney creates images. For example, asking for the “sound of waves crashing on the beach” would be text to sound. Voice cloning is a big element of AI audio. Another example would be an AI ‘watching’ a video and adding sound to it, like a foley artist would add footsteps or a creaking door to a movie scene. Lastly, AI audio can refer to microphones that only pick up certain speaker’s voices or headsets that cancel out all voices but your friends. It’s a small section of a broader category called multimodality. This week’s latest AI audio news: https://ethanbholland.com/category/ai/audio/
Autonomous Vehicles/Driverless Cars News of the Week: Driverless car news it often close to robot embodiment and multimodality (vision). I go with my gut each week around what to place in each category. My recommendation would be to follow Robotics/Embodiment and Multimodality also, as the fields are converging. Even AR/VR is part of driverless cars now, as robots are trained in simulations.
This week’s autonomous vehicle news: https://ethanbholland.com/category/ai/autonomous-vehicles/
Augmented and Virtual Reality (AR/VR) News of the Week: Augmented reality is when you see images or information overlaid on top of the real world, like a car windshield with a heads-up display, or glasses that can recognize faces and display names for people in view. Virtual reality is when you’re transported into another place, usually by wearing goggles (though a flight simulator could also be considered virtual reality).
In the past few months, VR has become one of the most important sectors in AI because simulations are now a core training ground for vision models, action models, robots, and driverless cars. Simulations can approximate physics across many virtual environments and enable reinforcement learning… where AI systems learn by interacting with a world through trial and error, without needing direct human instruction.
NVIDIA has been pushing this hard across its simulation and “physical AI” ecosystem (i.e. Cosmos). One of my favorite examples is a robotic dog trained in thousands of simulations to balance on a yoga ball, then transferred into the real world as “one more simulation,” interpreted through sensors. The robot didn’t “know” it was in the real world — and it balanced successfully without prior real-world practice.
Google also has a model called SIMA that can train agents to play video games and navigate simulated worlds — work that will eventually connect to embodied robots and driverless cars.
And AI pioneer Fei-Fei Li has a company dedicated to world models called World Labs. As of this latest update to my newsletter categories, they released a world model called Marble.
This is potentially the most important category other than Multimodality.
This week’s latest AR/VR news: https://ethanbholland.com/category/ai/augmented-reality-ar-vr/
Benchmarks News of the Week: Benchmarks are the scoreboards of AI. Standardized tests, leaderboards, and evaluations that help compare models (for coding, reasoning, image understanding, math, long-context, etc.). Benchmarks aren’t the whole story, but they’re a useful way to track progress.
This week’s latest benchmarks news: https://ethanbholland.com/category/ai/benchmarks/
Business/Enterprise News of the Week: This broad category is for stories that impact corporations and large scale AI implementation. Business includes funding rounds and acquisitions. Enterprise refers to a type of AI that is often custom built for a business or leverage an API to connect secure data to an AI model.
This week’s latest enterprise AI news: https://ethanbholland.com/category/ai/business-and-enterprise/
ByteDance News of the Week: ByteDance is best known for TikTok, but they also publish serious AI research and models (often in vision and multimodal). This is where I collect ByteDance-specific updates.
This week’s latest ByteDance news: https://ethanbholland.com/category/ai/bytedance/
Chips and Hardware AI News of the Week: Most of the chip news is NVIDIA usually, yet more and more Meta, Google, and OpenAI are starting toward their own manufacturing. This also includes data-centers and cloud-based computing.
This week’s latest chips and hardware news: https://ethanbholland.com/category/ai/chips-and-hardware/
Cohere News of the Week: Cohere is an enterprise-focused AI company. I use this category for Cohere-specific model releases, product updates, and anything related to how their tools get used inside businesses (especially search, retrieval, and “RAG”-style applications).
This week’s latest Cohere news: https://ethanbholland.com/category/ai/cohere/
Consumer Electronics AI News of the Week: This is a broad category meant to capture end user tools and products that incorporate artificial into their feature, from high-end grills to smartphones. Over time, I’m using this less and less, since AI is integrated into so many tools, but if something specific is released, I include it.
This week’s latest consumer AI news: https://ethanbholland.com/category/ai/consumer-electronics/
DeepSeek News of the Week: DeepSeek is one of the most important “open-weight” model families to track right now. This category is for DeepSeek-specific releases, benchmark performance, and signs of adoption, especially when DeepSeek starts showing up as the default model inside other tools, apps, and workflows. My January 28th cover says everything you need to know about DeepSeek.
This week’s latest DeepSeek news: https://ethanbholland.com/category/ai/deepseek/
Education AI News of the Week: There is a lot of buzz around the impact of AI in education. This section focuses both on the risks and rewards of how AI can impact learning. It’s broader than just K-12 and includes things like skills, trade, professional, and higher education. This is not about how to learn about AI, it’s about AI’s impact on learning.
This week’s latest education news: https://ethanbholland.com/category/ai/education/
Ethics/Legal/Security AI News of the Week: This section focuses on the impact AI is having on ethics (deep fakes, war, trust, false information, plagiarism, job loss, income), legal (rights, laws, regulations), and security (hacking, phishing, national interests, safety).
This week’s latest AI ethics/legal/security news: https://ethanbholland.com/category/ai/ethics-legal-security/
Figure News of the Week: Figure is one of the leading humanoid robotics companies. This category tracks Figure-specific updates… robot demos, real-world tasks, partnerships, and the “brains” that power the robots.
This week’s latest Figure news: https://ethanbholland.com/category/ai/figure/
Google AI News of the Week: Google is one of the biggest forces in modern AI. DeepMind research, Gemini, world simulations, driverless carts, TPU chips… and the downstream impact when those capabilities land in products like Search, Chrome, Android, Workspace, and YouTube. Individual Google stories will also be filed under the categories they match (images, audio, agents, robotics, etc).
This week’s latest Google AI news: https://ethanbholland.com/category/ai/google/
HuggingFace News of the Week: Hugging Face is basically the “GitHub + app store” for modern opens-source AI: a central hub where people publish models, datasets, and demo apps. If a model goes open-weight, a dataset goes viral, or a new tool becomes a default part of the AI ecosystem, there’s a good chance it shows up on Hugging Face first.
This week’s latest HuggingFace news: https://ethanbholland.com/category/ai/huggingface/
Imagery News of the Week: AI imagery covers “generative AI” image tools. This usually text-to-image, where a user enters a prompt (“a polar bear walking through NYC”) and a tool generates an image in the likeness of the description. This is different than AI vision, where an AI “looks at” an image and can derive context, details, and contents. AI vision is a subset of AI called multimodality. Imagery, in this case, is for image creation and modification/editing. Adobe’s AI tools would fall into this category. I’ll also include things like automatic masking, content editing, and object removal, even though that’s in between imagery and vision… but practically speaking it fits into images.
This week’s latest AI image news: https://ethanbholland.com/category/ai/images/
Inflection News of The Week: Inflection News of the Week: Inflection is an AI company best known for Pi, a personal assistant designed to feel more human-centered and conversational. As with other company categories, individual Inflection stories may also show up under the topic they match (agents, audio, etc.), but I use this category when the update is company-level or product-level.
This week’s latest Inflection news: https://ethanbholland.com/category/ai/inflection/
International AI News of the Week: A lot of international news will get cross listed in the chips, security, or open-source categories, however it’s nice to have a separate category for worldwide AI news. This is where French, Japanese, and Chinese models will usually appear.
This week’s latest international AI news: https://ethanbholland.com/category/ai/international/
Internet News of the Week: This category is about AI remaking the internet itself… the future of search, web browsers, SEO, web crawling, content discovery, paywalls, and the shift toward an “agentic web” where bots (not humans) navigate pages, click buttons, and complete tasks on our behalf. When AI changes how people find information (or skip websites entirely), it belongs here.
This week’s latest internet news: https://ethanbholland.com/category/internet/
Law News of the Week: This category is for AI-related law and regulation… copyright and IP fights, lawsuits, court rulings, licensing disputes, privacy, liability, antitrust, and new rules that shape what AI companies can (or can’t) do. There’s often overlap with Ethics and Security.
This week’s latest law news:
https://ethanbholland.com/category/ai/law/
Llama News of the Week: Llama is Meta’s open-weight model family, and it’s one of the biggest engines of the open ecosystem. This category tracks new Llama releases, licensing updates, fine-tunes, benchmark results, and the tools that grow up around it (local running, quantization, inference stacks, and “Llama-everywhere” deployment). It often overlaps with Open Source, Meta, and Local Models, but Llama is important enough to have its own category.
This week’s latest Llama news:
https://ethanbholland.com/category/ai/llama/
Locally Run AI Models News of the Week: This is a niche mostly for serious AI followers. It refers to AI that can be privately downloaded and run on a device without an internet connection. These have an array of powerful implications, from privacy to ethics to practical uses like Apple running a full AI on your phone, to corporate installations for security, to embodied robots with AI running in their virtual brain. Most experts believe the future of AI will be a combination of local and cloud based AI, working together.
This week’s latest locally run AI news: https://ethanbholland.com/category/ai/locally-run/
Manus News of the Week: Manus is a “full-stack” autonomous agent platform, less about inventing a new base model, and more about orchestrating tools and workflows on top of existing models to get real work done (research, coding, building apps, multi-step online tasks). It’s a good example of where the agent world is heading: the execution layer matters as much as the model. Recently, Meta announced it will acquire Manus, a reminder that the big platforms are racing to own agent capabilities, not just chatbots. I might stop posting to this category depending on what Meta does.
This week’s latest Manus news: https://ethanbholland.com/category/ai/manus/
Meta AI News of the Week: Meta AI News of the Week: This is a space for Meta-specific AI updates, especially Llama and the open-weight ecosystem Meta has helped catalyze, plus whatever Meta ships inside Instagram, WhatsApp, Facebook, and its creator/ad products. A lot of Meta stories also end up in Open Source and Agents, but I keep this category as a “one stop” view of Meta’s overall direction… especially huge hiring bonus and data center plans.
This week’s Meta AI news:
https://ethanbholland.com/category/ai/meta/
Microsoft AI News of the Week: This is a space for Microsoft-specific AI updates. Copilot across Windows and Office, Azure AI infrastructure, model releases like Phi, and Microsoft’s broader strategy around enterprise AI. Many individual stories will also appear in the topic categories they match (agents, security, chips, etc.), but this is the company-level catch-all.
This week’s Microsoft AI news: https://ethanbholland.com/category/ai/microsoft/
Mistral News of the Week: Mistral is one of the most important non–U.S. AI labs (France), and it has played a big role in the open-weight ecosystem. I use this category for Mistral-specific releases (models, capabilities, licensing), plus their product direction when it’s bigger than a single story.
This week’s latest Mistral news: https://ethanbholland.com/category/ai/mistral/
Mobile AI News of the Week: Mobile overlaps with almost everything (Apple vs. Google, chips, locally run models, consumer products), so I created a dedicated lane to keep “AI on phones” in one place. This includes on-device models, AI features baked into iOS/Android, and the practical reality of what people can actually run in their pocket.
This week’s latest mobile AI news: https://ethanbholland.com/category/ai/mobile/
Moonshot News of the Week: Moonshot AI is the company behind the AI model Kimi, and it’s earned its own category because it keeps showing up in serious model and agent conversations. This is where I collect Moonshot/Kimi-specific releases, agent features, and the ecosystem growing up around them.
This week’s latest Moonshot news: https://ethanbholland.com/category/ai/moonshot/
Multimodal AI News of the Week: This is the broad category for models that can work across more than one modality, text, images, audio, and video, and sometimes actions/tool use layered on top. Individual multimodal stories will often be filed under Images, Audio, Video, Robotics, AR/VR, etc., but this category is for the moments when the “one model that can do it all” trend is the story… especially vision and robotics. With segmentation, depth, and world simulation gaining momentum, I often cross-link between Multimodality, AR/VR, autonomous vehicles, and robotics/embodiment.
This week’s multimodal AI news: https://ethanbholland.com/category/ai/multimodal/
NVIDIA News of the Week: NVIDIA is the “picks and shovels” company of the AI era, the GPUs and software stack that power a huge portion of modern model training and inference. But NVIDIA is also increasingly a platform company: CUDA, networking, and now a growing ecosystem aimed at robotics and autonomy.
Lately, NVIDIA’s big push is “physical AI,” where simulations and world models generate synthetic data and training environments for robots and self-driving systems. Their Cosmos platform is part of that, a world-model/simulation stack designed to accelerate robotics and autonomous vehicle development. NVIDIA is also importantly the current U.S. open source leader (in my opinion).
This week’s latest NVIDIA news: https://ethanbholland.com/category/ai/nvidia/
OpenAI: OpenAI is perhaps the biggest brand in AI and one of the central forces in the current AI era. This category is for OpenAI-specific releases, research, policy moves, partnerships, and product direction. OpenAI stories will often overlap with other categories (Images, Audio, Agents, Ethics, etc.).
This week’s latest OpenAI news: https://ethanbholland.com/category/ai/openai-ai/
Open Source Models: “Open” usually means AI models that can be downloaded and run locally, fine-tuned, and built into other tools, without relying on a single vendor’s hosted API. (Sometimes the term is “open weights,” because the model weights are shared even if everything isn’t fully open source.) Open models can democratize access and speed up innovation, but they also reduce guardrails, which is why this category can feel like the wild west, depending on who you ask.
This week’s latest open model news: https://ethanbholland.com/category/ai/open-source/
Perplexity News of the Week: Perplexity is one of the leading “AI search” product, a research-first interface that answers questions with citations. Perplexity also has agents and an AI web browser calle Comet. Perplexity has been a flashpoint in the publishing debate: it’s faced plagiarism accusations and legal conflict with publishers, and it responded by rolling out publisher partnerships and revenue sharing. Regardless of where you land on that debate, Perplexity is one of the clearest bellwethers for how AI changes search, traffic, and attribution.
This week’s latest Perplexity news: https://ethanbholland.com/category/ai/perplexity/
Podcast/YouTube Clips and Op-Eds of the Week: This is for more general interviews and explainer videos and podcasts that provide access to leadership, demos of new products, and walkthroughs and tutorials. Videos focused on specific topics will live in the topic category (i.e. images), but broader videos will live here. This also includes opinion articles and commentary.
This week’s latest podcasts and YouTube clips: https://ethanbholland.com/category/ai/podcasts-youtube/
Publishing AI News of the Week: This category is about AI’s impact on publishing and the open web…copyright and licensing, crawling, SEO, paywalls, the erosion (or reinvention) of page views, and even the future of browsers as agents start consuming the internet for us. It’s similar to the future of the internet category, but focused on revenue models, IP, and content.
This week’s latest publishing news: https://ethanbholland.com/category/ai/publishing/
Qwen News of the Week: Qwen is Alibaba’s flagship model family, and it has become one of the most important open-model ecosystems to track — especially across multimodality, coding, and vision. When Qwen drops a big new model wave, it often ripples across the entire open-source and local-running community fast. Alibaba has its own category, but it’s been eaten by and crosslinks mostly to Qwen now.
This week’s latest Qwen news:
https://ethanbholland.com/category/ai/qwen/
RAG (Retrieval-Augmented Generation) News of the Week: RAG is one way to connect a language model to outside knowledge… your documents, a database, a website, a PDF library… so the model can retrieve relevant context before answering. RAG is the backbone of many enterprise AI tools because it keeps answers grounded in source material. RAG used to be notorious for hallucinations and brittle pipelines, but the tooling and techniques keep improving, especially as RAG merges with agents that can search, cite, and cross-check.
This week’s latest RAG news:
https://ethanbholland.com/category/ai/rag/
Robots / Robotics Embodiment News of the Week: This category is the collision of AI and the physical world…humanoids, robot dogs, warehouse robots, robot arms, and all the “brains” that make them useful (vision, planning, tool use, and reinforcement learning). It often overlaps with Autonomous Vehicles, Multimodality, and AR/VR, because simulation and world models are increasingly how these systems learn.
I first learned about embodiment from Alan Thompson in May 2023. I highly recommend his video explainer:
Also a must see video is Dr. Jim Fan’s TED Talk or speech on the robotic Turning test.
This week’s latest robotics/embodiment news: https://ethanbholland.com/category/ai/robotics-embodiment/
Safe Superintelligence Inc. News of the Week: Individual company products will often be placed in the categories they match (image, audio, agents, robots, etc). Occasionally, I’ll dedicate space to a company’s news if it’s broad or a major product release. Ilya Sutskever, former chief scientist and co-founder of OpenAI, opened this new AI-focused research lab in June 2024. In a bid to prioritize safety over commercial pressures, Sutskever’s startup concentrates solely on developing safe AI, insulating itself from the competitive pressures typical of tech companies.
This week’s latest Safe Superintelligence, Inc news: https://ethanbholland.com/category/ai/safe-superintelligence/
Science and Medicine News of the Week: This category tracks AI in scientific discovery and healthcare, things like drug discovery, protein and molecule modeling, lab automation, diagnostics, clinical workflows, and the broader trend of “AI that can do science.” It’s one of the most hopeful and exciting aspects in the whole field.
This week’s latest science and medicine news: https://ethanbholland.com/category/ai/science-and-medicine/
Security News of the Week: AI is now a force multiplier on both sides of cybersecurity and international relations/tensions. Defenders use it for vulnerability discovery, code review, incident response, and red teaming. Attackers use it for phishing, social engineering, malware development, and scale. This section is for the security-specific arms race as well as actual military and national security.
This week’s latest security news: https://ethanbholland.com/category/ai/security/
Technical and Developer Papers News of the Week: Everything that is too technical for general consumption goes here. These are stories I think are important, but might be inaccessible and confusing. Most often these are technical papers. It’s also a space for developer news and deep dives into how AI works, under the hood.
This week’s technical and dev AI news: https://ethanbholland.com/category/ai/technical-and-dev/
Video News of the Week: This category tracks AI video creation and editing… text-to-video, image-to-video, inpainting, style transforms, motion consistency, lip sync, upscaling, and the growing set of tools that can generate or modify real-looking footage. It also includes practical workflows for marketing, social video, and production. Often it dovetails with AR/VR, multimodality, and robotics.
This week’s latest video news:
https://ethanbholland.com/category/ai/video/
X/Twitter/Grok/: Grok is one of several AI’s developed by X, and it’s a bit blended in with Telsa and other Elon Musk technology, even robots and autonomous cars. Not every week will have a Grok section, but like Meta, Google, Apple, and OpenAI, X.ai will be in the news enough to have its own section. Optimus will also be in the robotic embodiment section and Telsa will be in the autonomous vehicles section.
This week’s latest X news: https://ethanbholland.com/category/ai/x/
Zhipu AI (Z.ai / GLM) News of the Week: This category tracks Zhipu AI and its GLM model family… releases, benchmarks, developer tooling, and adoption. It’s one of the major China-based labs showing up more and more in “frontier” conversations, especially around coding and agentic workflows.
This week’s latest Zhipu AI news:
https://ethanbholland.com/category/ai/zai/