Catalyst with Shayle Kann

• Inside a $300m bet on AI for materials discovery

  A big problem with using artificial intelligence to discover new materials? It struggles to predict beyond its training data. That means AI might be better at optimizing known materials than discovering entirely new ones — like a room temperature superconductor or carbon-capture sorbents.  But since we last covered the topic in September 2024, a few things have changed. OpenAI released its powerful O1 reasoning model. Large language models have also gotten better at math, physics, and coding. And lab automation — robots mixing liquids and powders, running characterization tests — has improved, allowing for a higher volume of experiments.  So, can these improvements overcome AI’s training data problem? In this episode, Shayle talks to Ekin Dogus Cubuk, cofounder of Periodic Labs, which raised $300 million seed round in September. Last year, Dogus took a more cautious view on using AI for materials discovery. Now though, he’s convinced there’s a clearer path forward. Shayle and Dogus cover topics like: Creating experimental and synthetic data to overcome AI’s limitations of predicting beyond its training set Why we should focus on breakthrough discoveries over easier, incremental wins The different roles humans and AI play in the discovery process Period’s focus on automated experimental labs using AI-generated hypotheses Resources: Catalyst: Can AI revolutionize materials discovery?   Latitude Media: This ‘superintelligence platform’ just raised $200m in seed funding   Latitude Media: Can AI get us closer to fusion?   The New York Times: Top A.I. Researchers Leave OpenAI, Google and Meta for New Start-Up Credits: Hosted by Shayle Kann. Produced and edited by Daniel Woldorff. Original music and engineering by Sean Marquand. Stephen Lacey is our executive editor.  Catalyst is brought to you by EnergyHub. EnergyHub helps utilities build next-generation virtual power plants that unlock reliable flexibility at every level of the grid. See how EnergyHub helps unlock the power of flexibility at scale, and deliver more value through cross-DER dispatch with their leading Edge DERMS platform, by visiting energyhub.com. Catalyst is brought to you by Bloom Energy. AI data centers can’t wait years for grid power—and with Bloom Energy’s fuel cells, they don’t have to. Bloom Energy delivers affordable, always-on, ultra-reliable onsite power, built for chipmakers, hyperscalers, and data center leaders looking to power their operations at AI speed. Learn more by visiting BloomEnergy.com. Catalyst is supported by Third Way. Third Way’s new PACE study surveyed over 200 clean energy professionals to pinpoint the non-cost barriers delaying clean energy deployment today and offers practical solutions to help get projects over the finish line. Read Third Way's full report, and learn more about their PACE initiative, at ⁠www.thirdway.org/pace⁠.

  --------  

  36:26

  --------

  36:26
• Unpacking DOE's proposal to transform data center interconnection

  Last Thursday, Energy Secretary Chris Wright directed the Federal Energy Regulatory Commission to consider rulemaking to fast-track interconnection for large loads — as long as they agree to be curtailable or colocate with dispatchable generation. So what does this proposal actually mean for interconnection?  In this episode, Shayle talks with Allison Clements, former FERC commissioner and current partner with digital infrastructure advisory firm ASG. Allison is also principal of 804 Advisory. Shayle also talks with Tyler Norris, doctoral student at Duke University’s Nicholas School of the Environment. Allison, Tyler, and Shayle cover topics like: How the proposal would standardize interconnection procedures for certain large loads, with study periods no longer than 60 days  The jurisdictional shift: asserting federal authority over a process traditionally under state purview   The types of eligible loads, including traditional data centers as well as ones that colocate with generation, also known as “hybrid facilities” The duration of flexibility and whether 2-hour, 4-hour, or longer durations are needed for curtailment Whether flexibility resources should be behind-the-meter or front-of-meter The potential disadvantages for bring-your-own-supply or bring-your-own-VPP Resources: Latitude Media: Wright directs FERC to fast track large load interconnection   Latitude Media: How the world’s first flexible AI factory will work in tandem with the grid   Latitude Media: OpenAI pushes the White House to invest in the grid to compete with China   E3: Demand Response as a Capacity Resource in SPP’s Era of Data Center Growth  Canary Media: In a first, a data center is using a big battery to get online faster  Credits: Hosted by Shayle Kann. Produced and edited by Daniel Woldorff. Original music and engineering by Sean Marquand. Stephen Lacey is our executive editor.  Catalyst is brought to you by EnergyHub. EnergyHub helps utilities build next-generation virtual power plants that unlock reliable flexibility at every level of the grid. See how EnergyHub helps unlock the power of flexibility at scale, and deliver more value through cross-DER dispatch with their leading Edge DERMS platform, by visiting energyhub.com. Catalyst is brought to you by Bloom Energy. AI data centers can’t wait years for grid power—and with Bloom Energy’s fuel cells, they don’t have to. Bloom Energy delivers affordable, always-on, ultra-reliable onsite power, built for chipmakers, hyperscalers, and data center leaders looking to power their operations at AI speed. Learn more by visiting⁠ ⁠⁠BloomEnergy.com⁠.

  --------  

  40:37

  --------

  40:37
• Five big questions about the future of energy

  We’ve covered AI’s massive power appetite in depth over the past year – with good reason. It’s the driving force behind much of the change and uncertainty in the energy world right now, from the error bars around our demand for electricity to the lineup of technologies vying to meet that demand.  In this episode Shayle talks to his colleague Andy Lubershane, head of research and partner at Energy Impact Partners, about five big questions arising in this uncertain load-growth environment. They cover topics like: The underappreciated factors that could flip the supply crunch to oversupply, like algorithmic efficiency gains, on-device inference, and off-grid data centers The winners of the AI-drive power boom, including utilities and grid equipment suppliers, and the potential losers like industry that relies on cheap power Whether there will be a “Cambrian explosion” or consolidation of nuclear reactors designs The prospects for enhanced geothermal after Fervo’s Cape Station comes online The future of grid-enhancing technologies like advanced conductors and dynamic line ratings, and whether they will make it out of “utility pilot hell” Resources: Steel for Fuel: Why does nobody know how much energy AI will consume? Open Circuit: How do we know if we’re in an AI bubble?   Catalyst: The US nuclear groundswell   Catalyst: How geothermal gets built   Latitude Media: In Georgia, stakeholders still can’t agree on data center load growth numbers   Credits: Hosted by Shayle Kann. Produced and edited by Daniel Woldorff. Original music and engineering by Sean Marquand. Stephen Lacey is our executive editor.  Catalyst is brought to you by EnergyHub. EnergyHub helps utilities build next-generation virtual power plants that unlock reliable flexibility at every level of the grid. See how EnergyHub helps unlock the power of flexibility at scale, and deliver more value through cross-DER dispatch with their leading Edge DERMS platform, by visiting energyhub.com. Catalyst is brought to you by Bloom Energy. AI data centers can’t wait years for grid power—and with Bloom Energy’s fuel cells, they don’t have to. Bloom Energy delivers affordable, always-on, ultra-reliable onsite power, built for chipmakers, hyperscalers, and data center leaders looking to power their operations at AI speed. Learn more by visiting BloomEnergy.com.

  --------  

  44:23

  --------

  44:23
• Frontier Forum: The new power map for AI infrastructure

  As AI reshapes the industrial landscape, companies are questioning whether the grid can keep pace. Permitting delays, transmission constraints, and reliability risks are forcing developers to rethink where power comes from. In this episode, KR Sridhar, CEO of Bloom Energy, lays out a radically different vision. He believes that many data centers will ultimately operate like refineries — powered by captive, off-grid generation that prioritizes resilience, speed, and local control over traditional grid economics. Sridhar argues that solid-state fuel cells have become an ideal solution to meet data center needs at AI speed and scale. They can be deployed in months rather than years, follow digital loads in real time, and integrate with future zero-carbon fuels like hydrogen. “I truly believe that this is a cyclical trend that’s going to continue for well over a decade,” said Sridhar. This episode features an edited version of our live Frontier Forum conversation about what a future-proof AI power strategy really looks like. We talk about the tension between off-grid and grid-connected approaches, the importance of speed to power, carbon capture, and supply chains over the next decade of growth. The conversation also touches on Bloom’s new white paper, Fuel Cells: A Technology Whose Time Has Come, which argues that onsite generation can deliver AI-scale reliability and lower emissions. You can watch the full Frontier Forum conversation with audience Q&A here.

  --------  

  36:04

  --------

  36:04
• Calibrating hype with Akshat Rathi

  In the climate space, every idea sits somewhere along the hype continuum. Some command outsize attention. Others fly under the radar despite big potential. And a rare few hit the sweet spot, earning exactly the buzz they deserve. But how do you tell which is which? In this episode, Shayle teams up with Akshat Rathi, senior reporter for climate at Bloomberg News and host of the Zero podcast, to sort it out. Akshat and Shayle run through a list of hot topics and place each one on the hype continuum. They cover topics like: Using DERs to meet load growth Co-locating generation with data centers Infrastructure bottlenecks like generation, transmission, and transformers The roles of venture capital and the Paris Agreement in shaping markets A grab-bag of other topics like sodium-ion, advanced geothermal, and advanced nuclear Resources: Catalyst: The new wave of DERs  Catalyst: When to colocate data centers with generation   Zero: The Device Throttling Our Electrified Future Zero: The Gas Turbine Shortage Might Be a Climate Problem   Credits: Hosted by Shayle Kann. Produced and edited by Daniel Woldorff. Original music and engineering by Sean Marquand. Stephen Lacey is our executive editor.  Catalyst is brought to you by EnergyHub. EnergyHub helps utilities build next-generation virtual power plants that unlock reliable flexibility at every level of the grid. See how EnergyHub helps unlock the power of flexibility at scale, and deliver more value through cross-DER dispatch with their leading Edge DERMS platform, by visiting energyhub.com. Catalyst is brought to you by Bloom Energy. AI data centers can’t wait years for grid power—and with Bloom Energy’s fuel cells, they don’t have to. Bloom Energy delivers affordable, always-on, ultra-reliable onsite power, built for chipmakers, hyperscalers, and data center leaders looking to power their operations at AI speed. Learn more by visiting BloomEnergy.com.

  --------  

  41:14

  --------

  41:14

Fler podcasts i Näringsliv

Trendiga poddar i Näringsliv

Om Catalyst with Shayle Kann

Catalyst with Shayle Kann: Poddsändningar i Familj