标签：#ai

The elevated streaming experience is happening in collaboration with Prime Video.

"Maybe that's my responsibility, to live alone with the truth."

Buzz is a group chat platform for the workplace that puts humans and their AI agents in the same conversation.

Over the past decade, streaming platforms competed by dominating individual formats like music, video, podcasts, or audiobooks. Now, as AI makes it easier to create, organize, and recommend content, those distinctions are fading, pushing companies like Spotify, Netflix, YouTube, and TikTok to become all-purpose entertainment destinations instead.

Substack will now help users determine whether what they're reading may have been written by AI. A new tool coming to the platform can scan posts, notes, replies, and comments to provide an estimate of how much text could be AI-generated or written with AI assistance, according to a blog post published on Tuesday. The […]

Substack is partnering with Pangram to add an AI detection feature.

Most "AI agent" libraries fall into one of two buckets. Either they're a big framework you spend an afternoon configuring, or they're a tiny toy that drops the one feature you actually need in production: the ability to stop and ask a human before the agent does something you can't undo. I wanted the middle. So I wrote yieldagent : a small agent loop you can read end to end, with human-in-the-loop pause/resume built in, and no runtime dependencies. This post walks through how it works and why it's built the way it is. What an agent loop actually is Strip away the branding and an "agent" is a loop: Send the conversation to the model, along with the tools it's allowed to call. If the model asks to call a tool, run it and append the result. Repeat until the model answers without asking for a tool. That's it. The model decides the control flow at runtime; your job is to run the tools and feed the results back. Here's the core, lightly trimmed: for ( let step = 0 ; step < maxSteps ; step ++ ) { const reply = await call ( messages , toolSpecs ); messages . push ( reply ); if ( ! reply . tool_calls ?. length ) { yield { type : " final " , text : reply . content , messages }; return ; } for ( const tc of reply . tool_calls ) { const args = JSON . parse ( tc . function . arguments ); const result = await tools [ tc . function . name ]. run ( args ); messages . push ({ role : " tool " , tool_call_id : tc . id , content : JSON . stringify ( result ) }); } } Everything else in the library is in service of making this loop observable, testable, and safe to run against the real world. Why an async generator Notice the yield . The loop is an async generator, so the caller drives it: for await ( const step of agent ({ call , tools , messages })) { if ( step . type === " tool-start " ) console . log ( " -> " , step . tool , step . args ); if ( step . type === " final " ) console . log ( step . text ); } Every step (each tool call, each result, and the final answer) is handed back to

For a long time, I believed coding was only for people who studied computer science or worked as professional developers. Whenever I saw a screen filled with code, it looked like a completely different language. There were brackets, symbols, functions, and terms I did not understand. I assumed learning it would require years of study before I could create anything useful. That changed when I encountered a repetitive task at work. I was regularly copying information from a spreadsheet, checking each row, preparing an email, sending it, and updating the status manually. The work was manageable, but completing the same process repeatedly took time and left room for mistakes. I started wondering if the spreadsheet could do some of the work for me. That question led me to Google Apps Script. At first, I did not even know where to begin. I understood the result I wanted, but I did not know how to translate it into code. I could explain the process clearly to another person, but explaining it to a computer felt different. AI became my starting point. I described the task and asked it to create a script. Within seconds, it gave me several lines of code. I copied them, ran the script, and immediately received an error. My first reaction was frustration. I had expected the code to work because it looked complete. But I soon realized that generated code was not automatically working code. I went back and explained the error. AI suggested a change, so I tested it again. Another issue appeared. I repeated the process until the automation finally worked. The moment it worked, something changed in the way I viewed coding. I did not suddenly become a programmer, but I had created something useful. A task that previously required several manual steps could now happen automatically. I became curious about what else I could build. I started experimenting with confirmation emails, timestamps, form submissions, missing-data checks, and automatic reports. Each project introduced me to a

It started with one simple question: Why am I still doing this manually? I was working on a spreadsheet, checking information row by row, sending emails, and updating statuses. The process was not difficult, but it was repetitive. One small mistake could mean sending incorrect information, overlooking a request, or forgetting to update a row. I knew there had to be an easier way. I wanted the spreadsheet to detect when a status changed to Sent , find the email address in the same row, send the correct message, and add a timestamp after the email was sent. The idea sounded simple in my head. The problem was that I did not know how to build it. I was still learning how to code, so I asked AI for help. My first prompt was something like: Create a script that sends an email from Google Sheets. AI immediately generated a script. It looked impressive, but it did not work the way I expected. The script checked the wrong column, used information from the wrong cells, and failed when some required details were missing. That was when I realized the problem was not only the code. My instructions were too vague. I tried again, but this time I described the entire process: When the status in Column G changes to "Sent," get the email address from Column E and send a confirmation email. After the email is sent successfully, add a timestamp to Column H. Do not send the email if any required information is missing. The result was much closer to what I needed. When I changed the status to Sent and received the email automatically, I felt genuinely excited. It was only a small automation, but it removed several manual steps from the process. After building that first workflow, I started noticing repetitive tasks everywhere. A form submission could automatically send a confirmation email. A spreadsheet could detect missing information before processing a request. A completed action could record the date and time. Reports could be organized without manually copying every row. Tasks that

An app called ToneAdapt is pulling in roughly $25K/month, five months after launch, built by one person with no funding and no team. It solves one narrow problem: guitarists want to sound like their favorite recordings, but tutorials and gear reviews rarely match what's actually sitting in their room. Input your guitar, amp, and pedals, pick a song, and it spits out the exact settings to get there. It's not a big, category-defining idea. It's a small, sharp one, solved well, and marketed relentlessly. That combination is worth breaking down. The Numbers The founder shared these publicly: combined web and mobile revenue is running around $25K/month, split roughly evenly between a Stripe-powered website and a mobile app on RevenueCat. Mobile has under 400 active subscribers. Pricing is aggressive for a mobile app, $10/week, or about $60/year, which works because it's solving a moment of real frustration rather than sitting as a passive subscription people forget about. Worth noting: these numbers come from a founder interview, so treat them as a snapshot of where the business was at that point rather than a live dashboard. The shape of the story, fast growth on a narrow niche with aggressive pricing, is the part worth paying attention to. Marketing: Almost Entirely UGC The founder posted three times a day across Instagram, TikTok, YouTube, and Facebook. Most of it didn't land. Once a specific format started converting, he stopped experimenting and went all-in, bringing in UGC creators and running paid ads on top of the format that already worked. That's the actual playbook: post relentlessly, watch for the one format that converts, then stop spreading effort thin and pour it all into what's already working. The Pattern Behind It The reason this works isn't guitar tone specifically. It's a translation problem: someone sees a result they want, but the instructions that got that result assume gear they don't have. The gap between "here's what worked for them" and "here's

Everyone is arguing about AI ROI right now. Boards want a number. Consultants are selling frameworks. And the headlines look brutal: an MIT report claimed 95% of enterprise GenAI pilots showed no measurable return, and a Forbes piece this January says 56% of CEOs see zero ROI from AI. So AI is a bust, right? Here is the thing. Ask any of those same companies to turn off email for a week. Just try it. Nobody ever ran a six month ROI study on email. Nobody had to. It became the way work happens, and the return stopped being a line item because it was everywhere. I think AI is following the same path, and the data backs it up. The gap between the pilot and the person That same MIT report has a stat that got way less coverage than the 95% number: 90% of employees regularly use personal AI tools for work, while only 40% of companies have an official LLM subscription. Workers adopted it at more than twice the rate of their own employers. The report even found that 70% of workers prefer AI over a colleague for quick stuff like drafting emails and basic analysis. So the official pilot fails its KPI review while the people inside the building are quietly using AI multiple times a day. That is not a failed technology. That is a measurement problem. Worth noting: the 95% figure itself has been heavily criticized. It came from 52 interviews and a narrow definition of success. I would not build a strategy on that number in either direction. Why the ROI is invisible The value seems to be landing in the same place email's value landed: the unglamorous middle of the workday. Drafting the reply. Summarizing the thread. Turning messy notes into something you can send. None of that shows up as a new revenue line. It shows up as an hour you got back and immediately spent on something else. You cannot easily measure that, but you can feel it. Take the tools away from a developer who has been using them for a year and watch what happens. People cannot imagine working without it anymore.

New data centers built through 2033 could consume as much electricity as India uses today.

Anthropic blocks authors from opting out of $1.5B settlement at last minute.

Google released Gemini 3.6 Flash, 3.5 Flash-Lite, and Flash Cyber, but the continued absence of Gemini 3.5 Pro raises fresh questions about its AI strategy.

OpenAI launches the ChatGPT for Small Businesses program, helping entrepreneurs build AI skills, automate work, and grow with ChatGPT Work.

Twitch is giving parents more control over how their children are using the streaming platform, including the ability to block them from broadcasting entirely. Parental controls are now available that allow guardians to link Twitch accounts with their 13- to 17-year-old children, providing account management features and a weekly email summarizing their teen's activity, including […]

There are new 3.6 and 3.5 models today, but Google is already training Gemini 4.

A federal judge has signed off on Anthropic's $1.5 billion class action settlement with authors who accused the company of training its AI models on copyrighted books, as reported earlier by Reuters. In an order on Monday, Judge Araceli Martínez-Olguín writes that the settlement will provide "meaningful relief," offering authors around $3,000 for each book […]

Canvases turn AI into interactive workspaces where you can visualize information, explore workflows, and take action across complex tasks. The post How to build interactive experiences with canvases appeared first on The GitHub Blog .

Open-source LLMs stopped being the budget option in 2026. Kimi K3 sits level with Claude Opus 4.8 on the Artificial Analysis Intelligence Index (its hosted API is live; the weights themselves are expected by July 27), GLM-5.2 held the top open-model spot before it, and the field behind them is deep enough that the hard part is choosing. This ranking covers the nine best open-weight models right now — on license, context window, hardware reality, and the per-token price you actually pay. Every one of them is available through LLM Gateway with one key, at each provider's published rate, so you can A/B any two of them by changing one word in a request. 1. Kimi K3 — the open frontier Moonshot AI · 2.8T params · 1M context · $3.00 / $15.00 per M The largest open-weight model ever announced — with one caveat: the weights are not downloadable yet. Moonshot expects to release them by July 27, 2026, and the license is still unannounced; the hosted API has been live since July 16. Ranks 4th of 189 models on the Artificial Analysis Intelligence Index — tied with Claude Opus 4.8 and GPT-5.5 — and took first place in Arena's blind Frontend Code testing. Always-on reasoning, vision, tools, and output configurable up to 1M tokens. The open model to beat, priced accordingly. Full breakdown here . Best for: teams that want closed-frontier quality with open-weight freedom. 2. GLM-5.2 — the value flagship Z.ai · 744B params · 1M context · $1.40 / $4.40 per M MIT-licensed, weights on Hugging Face, and the top-ranked open model until K3 arrived. A real 1M-token context, strong agentic-coding results, and built-in web search support — with output at under a third of K3's rate and input at about half. Also the largest model on this list that fits a single 8-GPU node (or one 512 GB Mac Studio) at INT4. Best for: the best capability-per-dollar in the open field. 3. DeepSeek V4 Pro — frontier scale at commodity prices DeepSeek · 1.6T params (49B active) · 1M context · $0.435 / $0.87 per M MI