Podcast anything.

Any source. One RSS feed.

Blogs

X accounts

Hacker News

Podcasts

Any website

The Daily FMSummarizes & narrates

One RSS feed

Subscribe to ready-made pods

Grab any pod from the library with its RSS feed (no account needed!)

Combine pods into one feed

Merge many podcasts into your own custom pod to get them all in a single feed

Customize your own pod

Create a custom summary pod with your own sources

Sources can be the things you already follow:

Websites: any article, blog, or newsletter link
X: the latest posts from your favorite @handles
Hacker News: top stories and their comment threads
Podcasts: fresh-episode recaps of any show
State legislatures: weekly bill activity

Connect your AI agent

Connect it over MCP and it can build and run pods for you

Featured Pods

AI Daily

anthropicopenaideepmindblognewssimonwillison@karpathy@sama@AnthropicAI@swyx@levelsio@yoheinakajima@dair_ai@rowancheung@rasbt@bcherny@steipete

Combine these episodes into your pod

All episodes from this podcast will be fed into your own.

AI Daily August 14: OpenAI GPT-5.6 Sol Hits 750 Tokens/Second; Google Launches Gemini 3.7 FlashHere is today's AI Daily for Friday August 14th. Yesterday, OpenAI previewed Ultrafast mode for GPT-5.6 Sol, powered by Cerebras. The company says select API customers can get speeds of up to 750 tokens per second—roughly 14 times standard mode. That matters most for work where delay breaks the experience: live voice, customer support, interactive coding, trading, and security operations. Faster generation alone does not make an agent faster, since tool calls and external systems still take time. But it pushes the bottleneck away from the model and toward the rest of the workflow. Google also released Gemini 3.7 Flash, just weeks after 3.6 Flash. It is positioned as a broadly capable, lower-cost workhorse for coding, web development, knowledge work, and agents. Google is offering an introductory 50 percent price reduction through year-end, while early third-party measurements report improvements in coding and agent benchmarks. The rollout is unusually broad: Gemini 3.7 Flash is already reaching the API, AI Studio, Android Studio, Gemini Enterprise, and outside developer tools. The competitive battlefield is clearly shifting toward models that are not only capable, but cheap and easy to put into production. Meanwhile, DeepSeek open-sourced DeepSeek Harness in developer preview. Rather than being another coding assistant, it is a plugin-based runtime for long-running agent tasks, with visible trajectories and cache-conscious history management. It is explicitly unstable, but the release highlights a growing consensus: the harness—the system controlling memory, tools, permissions, and recovery—may matter as much as the underlying model. Finally, new research is raising flags about agent reliability. One study finds that context-compaction systems retain only 17 percent of persistent user constraints, such as “do not delete emails without confirmation.” Another concludes that differences among agents explain under 3 percent of variance in several popular agent benchmarks. In other words, rankings may look more precise than they are. The takeaway: raw model intelligence is improving quickly, but real-world advantage increasingly comes from speed, cost, robust agent infrastructure, and evaluations that measure actions—not just answers. Thank you for listening to AI Daily from The Daily FM. See you tomorrow!

Financial Markets

bloombergcnbcfinanceft@WSJmarkets@markets@business@CNBCnow@ReutersBiz

Combine these episodes into your pod

All episodes from this podcast will be fed into your own.

Financial Markets August 14: Silver Lake Eyes Workday Buyout as Apple Builds China AI ModelHere is today's Financial Markets for Friday August 14th. Yesterday’s softer producer-price data helped lift the S&P 500 to another record close, extending a rally built on two straight days of benign inflation readings. Technology led again, with Sandisk among the notable gainers, while investors further reduced expectations that the Federal Reserve will raise rates at its September meeting. This morning, however, U.S. futures are slipping, suggesting some end-of-week caution after the run to record highs. Gold is also headed for a weekly loss despite the softer inflation backdrop, a sign that investors remain willing to hold risk assets. The biggest corporate development is a potential blockbuster software deal. Reuters reports that Silver Lake is in talks to acquire Workday, in what could rank among the largest software buyouts ever. The talks matter beyond Workday: they suggest private equity sees long-term value in established enterprise-software businesses, even as investors debate whether AI agents could disrupt the sector. A completed deal could also encourage more take-private activity among companies whose public valuations have lagged their strategic value. AI remains central to global technology competition. Reuters says Apple has trained a large-language model specifically for China, with Alibaba’s support. That is a departure from Apple’s prior reliance on third-party models for AI features in the country and underscores how major technology firms are adapting products, partnerships, and data strategies to navigate China’s regulatory environment. Trade policy is also reshaping a smaller but rapidly growing technology market. President Trump said the U.S. will impose tariffs on imported drones and components, including products from key allies, citing reliance on foreign supply. Domestic drone makers rose on the news, but the tariffs could raise costs for commercial users and manufacturers that depend on imported parts. Finally, energy remains a cross-market risk. Crude has moved higher amid renewed U.S. threats against Iran, even as broader oil signals remain mixed. The market’s message is constructive but selective: cooler inflation supports equities, while geopolitics, tariffs, and elevated valuations keep volatility close at hand. Thank you for listening to Financial Markets from The Daily FM. See you tomorrow!

Latent Space in 3 minutes

pod:api.substack.com

Combine these episodes into your pod

All episodes from this podcast will be fed into your own.

Latent Space in 3 minutes: 🔬The BioAI Phase Shift - Matthew McPartlon & Neil Patil, Chai DiscoveryHere is The Daily FM summary of the Latent Space that aired on Tuesday August 11th. This episode explored what Chai Discovery calls a “phase shift” in BioAI: moving drug discovery away from slow, brute-force experimentation and toward precision engineering with AI models. Chai cofounder Matt McPartlon and platform leader Neil Patil described the company as a software and modeling layer for pharma rather than a company trying to build its own drug pipeline. Their pitch is that better protein-structure and protein-design models can help drugmakers find therapeutic candidates faster, with more control over where and how they bind. Chai has partnered with Eli Lilly, Pfizer, Novartis, and argenx, and argues that its success is tied directly to whether those partners make better medicines. The central technical focus was antibodies: Y-shaped immune proteins whose tips can be engineered to attach to disease targets. Traditionally, antibody discovery often means immunizing mice or screening billions of possibilities in yeast-display experiments, then hoping a few molecules stick. Chai’s models aim to generate candidates intentionally, including molecules that bind a specific site, avoid similar proteins that could cause side effects, or bind both human and animal versions of a target to support preclinical testing. McPartlon described Chai 1 as a structure-prediction model: given a protein sequence, it predicts the 3D shape. Chai 2 made the bigger leap into design, generating both a candidate’s sequence and structure for a target. In a high-profile internal challenge, the team designed antibodies against 50 targets and got binders for roughly half. One remarkable validation result showed a predicted structure only 0.33 angstroms from cryo-EM measurement—about one-third the width of an atom. The team initially suspected the lab had accidentally sent their own prediction back. But the guests stressed that binding is only the beginning. Good drugs also need strong affinity, selectivity, stability, safety, manufacturability, and the ability to avoid unwanted aggregation. Chai 3 and subsequent iterations focus on pushing toward therapeutic-grade molecules rather than treating hit discovery and lead optimization as separate, years-long waterfall stages. Patil argued that the product should not look like a chatbot. Instead, Chai built something closer to Figma, SolidWorks, or Photoshop for molecules: scientists can visually select an epitope—the precise target region—and ask the model to generate binders under constraints. The product must also meet pharma’s strict security and intellectual-property requirements, including isolated customer deployments. A moving moment came when Patil recalled a pharma scientist crying after Chai helped generate a binder for a target she had worked on unsuccessfully for ten years. That illustrated the broader claim: these systems are beginning to work in real campaigns, not just on benchmarks. The biggest remaining bottleneck, McPartlon said, is rapid experimental validation. Models can generate hypotheses quickly, but wet-lab proof still takes weeks or months. Patil’s answer was talent: BioAI needs more engineers and researchers to realize that biology is becoming computationally accessible. Their final message was ambitious but clear: as models predict structures within atomic accuracy and reliably design binders, biology may become less like feeling around in the dark and more like an engineering discipline. Thank you for listening to Latent Space in 3 minutes from The Daily FM. See you next time!

View the whole library →

Subscribe to one of these pods with the RSS feed, or create your own →