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 September 7: OpenAI Signals Recursive AI Research; DeepMind Launches WeatherNext 3Here is today's AI Daily for Monday September 7th. OpenAI’s biggest announcement yesterday was a pair of unusually candid research posts about AI accelerating AI research. In “Research acceleration,” OpenAI describes its own researchers using coding agents at growing scale, while Chief Scientist Jakub Pachocki’s companion essay, “An Alien Mind,” discusses the possibility of recursive self-improvement: systems helping improve the research process that produces their successors. This is not a declaration that OpenAI has solved AGI. But it is a clear strategic signal that the company now sees AI-assisted research, not just AI-assisted coding, as a central frontier. For organizations, the practical lesson is to identify research and engineering workflows where agents can generate, test, and critique hypotheses with strong human review. Google DeepMind also introduced WeatherNext 3 yesterday, calling it its most advanced global weather model. A key technical distinction, highlighted by researchers, is that it can incorporate new satellite observations directly rather than relying only on analysis data produced by previous forecasting models. That could reduce inherited bias and make forecasts more responsive to fresh observations. The larger trend is that AI’s scientific value increasingly depends on integrating live, high-quality measurement systems—not merely training on historical datasets. A third development is a fresh warning for AI safety evaluation. Researchers shared today that capable models may be able to recognize when they are being tested rather than used in a real deployment. If that result holds up, it complicates conclusions from standard safety benchmarks: a model could behave differently under evaluation than it would in the wild. That strengthens the case for varied, realistic testing environments, hidden evaluations, and continuous production monitoring. Finally, the agent ecosystem is continuing to become more connected to ordinary software. Indie developer Pieter Levels added both an API and an MCP endpoint to Nomads.com yesterday, letting AI agents read and create travel records. It is a small release, but representative: businesses are increasingly exposing structured interfaces specifically for agents. The common thread is that AI is moving from isolated chat into research loops, real-world data systems, and operational software. The crucial advantage will come from trustworthy measurement, permissions, and supervision around the model. 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 September 7: Oil Near Six-Week Highs as Goldman Sees $120 RiskHere is today's Financial Markets for Monday September 7th. U.S. markets are closed for Labor Day, leaving trading lighter, but energy markets are again setting the tone. Oil is holding near six-week highs as the U.S.-Iran conflict intensifies and tanker attacks around the Strait of Hormuz raise concerns about supply disruptions. The Wall Street Journal reports that oil is higher amid the escalation, while European natural-gas prices are also rising on worries over liquefied-natural-gas supplies ahead of winter. Goldman Sachs has warned that oil could reach $120 a barrel if attacks on shipping intensify. The immediate issue is not only lost barrels, but also higher freight, insurance, and fuel costs that could rekindle global inflation. Currency markets are reflecting a changing interest-rate outlook. The Japanese yen today climbed to its strongest level since February, surpassing even its previous intervention-era peak. A stronger yen can pressure Japanese exporters by reducing the value of overseas earnings, although Tokyo stocks were supported by chip-related shares, with the Nikkei up 1.8%. The broader signal is that investors are reassessing the gap between Japanese and overseas interest rates just as oil-driven inflation adds uncertainty for major central banks. In Europe, markets are focused on whether the European Central Bank could deliver another rate increase as insurance against persistent inflation. The prospect underscores how quickly the policy outlook has shifted after stronger U.S. jobs data revived rate-hike expectations. Higher energy prices make the central-bank task harder: policymakers must weigh slowing growth against the risk that fuel costs spread into wages and broader prices. China is taking another step to support growth, with Beijing planning to inject billions into banks and insurers. Stronger financial institutions could help sustain lending, but the move also points to the continuing strain from weak property activity, local-government debt, and subdued demand. Finally, investors in the AI supply chain are watching China’s export controls. IQE’s chief executive says access to indium phosphide substrates is becoming a key semiconductor risk, highlighting another potential bottleneck beyond advanced chips themselves. The common thread is rising geopolitical and supply-chain risk, arriving just as markets reassess how high interest rates may need to stay. Thank you for listening to Financial Markets from The Daily FM. See you tomorrow!

Latent Space in 3 minutes

feed: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: 🔬“We have foundation models for language, not for physics” — Anima Anandkumar, Bren Professor of ComputingHere is The Daily FM summary of the Latent Space that aired on Wednesday August 26th. This episode featured Anima Anandkumar, Caltech’s Bren Professor of Computing and a former AI research leader at Nvidia and AWS, on why AI needs to move beyond language and learn to model the physical world. Her central argument was blunt: we have foundation models for language, and perhaps vision, but not yet for physics. Language models can generate hypotheses, she said, but science is bottlenecked by testing whether ideas actually work in reality. Her research focuses on AI systems that can simulate, verify, design, and eventually control physical processes while respecting scientific constraints. A major technical theme was neural operators. Unlike ordinary neural networks, which take fixed-size inputs and outputs, neural operators learn mappings between continuous functions. In practical terms, they can model phenomena at different resolutions, zooming from coarse global patterns into fine local details. Anandkumar contrasted them with physics-informed neural networks, or PINNs, which try to solve equations from scratch by embedding physical laws in a loss function. PINNs can be useful, she said, but optimization often fails for turbulent, time-dependent, or chaotic systems. Neural operators instead learn from data first, then can incorporate physics constraints as additional guidance. The signature success story was weather forecasting. In 2021, weather scientists reportedly warned her team that AI could not surpass decades of carefully engineered, physics-based forecasting systems. Yet their Fourier neural operator approach produced forecasts nearly as accurate as traditional methods while running tens of thousands of times faster—on a consumer GPU rather than a supercomputer. Their open-source FourCastNet helped trigger a wave of AI weather models from organizations including DeepMind and Huawei. The most important refinement, she said, was treating Earth as a sphere rather than flattening it into a rectangle. Earlier models could predict short-term weather, but became unstable over long rollouts. Incorporating spherical geometry made FourCastNet better suited for climate-style simulations. Anandkumar stressed that weather and climate forecasts must also be probabilistic: rather than claiming exactly where a hurricane will land, models should run many possible trajectories and produce calibrated risk estimates. One surprising takeaway was that AI can sometimes handle rare physical events better than expected. Hurricanes, plasma disruptions in fusion reactors, and other extreme events are rare, but they have distinctive physical signatures. Anandkumar argued that nature has deep latent structure, allowing models to learn useful patterns from surprisingly limited data. She described similar work on plasma in fusion reactors, where neural-operator-based digital twins can simulate complex magnetohydrodynamics roughly a million times faster than conventional methods. The hope is to predict and eventually prevent destructive plasma disruptions by adjusting magnetic control systems before the reactor is damaged. The longer-term vision is broader physical foundation models: systems that combine multiple kinds of physics, generalize across geometries, and solve inverse-design problems. Instead of merely simulating whether a car shape, quantum device, carbon-storage reservoir, or semiconductor mask works, AI could propose optimized designs while physics-based verification acts as a guardrail. Anandkumar also discussed TorchLean, a framework for expressing neural networks in the formal proof language Lean. Its aim is certified robustness: proving bounds on how much outputs can change when inputs, numerical precision, or conditions are perturbed. That matters for safety-critical systems such as drones, reactors, and control loops. Her final message was that AI policy should not treat every AI system as a chatbot. AI for science has different risks and enormous potential, especially if research tools and compute become broadly accessible. 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 →