July 2026
The panel’s central dispute was whether Washington can make its security relationship with Israel more conditional—or phase down aid—without sacrificing the leverage and regional influence that a deeper break could forfeit. Dan Shapiro argued for a managed transition that preserves Israel’s deterrence and U.S. influence; Dahlia Scheindlin saw conditionality as an increasingly durable political response to Israeli policy, prolonged war and civilian harm. Ted Deutch countered that aid policy cannot be serious if it treats Israel as solely responsible for a conflict in which Hamas and other regional actors retain agency.
Former U.S. national security official Robert Greenway argues that U.S. military action has reduced Iran’s nuclear, missile and drone capabilities enough to justify eliminating its remaining ability to threaten regional trade and energy flows. Senator Chris Murphy counters that Tehran’s survival and continued leverage over the Strait of Hormuz have left Washington with a weaker hand in any future negotiation. Carnegie’s Karim Sadjadpour adds that a militarily diminished Iran may nonetheless emerge more confident that coercion works.
Jordanian Deputy Prime Minister and Foreign Minister Ayman Safadi argues that the region’s wars cannot be contained through ceasefires or military arrangements alone: durable security requires respect for state sovereignty, lawful access to shipping and humanitarian aid, and a political path to Palestinian statehood. In a conversation moderated by NBC’s Andrea Mitchell, Safadi says Jordan supports diplomacy with Iran, a gradual restoration of Lebanon’s state control over weapons, and Gaza aid access independent of stalled governance plans, while warning that Israeli settlement expansion is closing off the two-state solution.
Akif Çağatay Kılıç, Türkiye’s chief adviser on national security and foreign policy, argues that Ankara’s geography requires it to keep working relationships open across rival blocs, from NATO and the EU to Russia, Ukraine and Iran. He presents Türkiye as a venue for diplomacy rather than a power able to dictate settlements, while acknowledging that its own defense ties with the United States remain constrained by the unresolved dispute over its Russian S-400 system and access to F-35 aircraft.
Y Combinator president and CEO Garry Tan argues that AI’s largest productivity gains will come not from access to better models but from companies that build and maintain institutional memory for agents. His proposed “company brain” combines curated knowledge, context retrieval and reusable skill files so agents can act on prior work rather than repeatedly starting from scratch. The strategic asset, Tan says, is the organization’s accumulated procedures and judgment—not the model weights, which competitors can rent.
Thinking Machines Lab is positioning its first model, Inkling, as an open-weight system built for customer customization rather than as the industry’s benchmark leader, TBPN’s John Coogan argues. Led by former OpenAI technology chief Mira Murati, the company pairs the model with a fine-tuning service that could let customers retain control of the weights while paying Thinking Machines to adapt and operate them. Coogan and Tyler Cosgrove also question how cleanly Inkling can be described as free of distillation, given its reported use of synthetic data from another open-weight model.
The discussion casts technology projects as constrained by the systems they operate: Roblox controls the Robux economy through which creators earn and, if eligible, cash out, while its roughly 70/30 revenue split remains disputed. TSMC’s displayed $38 billion 2024 capex estimate shows the capital required to expand advanced-chip capacity, and California Forever’s proposed Solano County city still depends on local approval despite its land purchases.
OpenAI’s Dominik Kundel argues that ChatGPT’s GPT-5.6 desktop app is designed to complete work in the systems where it already resides, rather than merely generate instructions or text in a separate chat window. He divides that work by environment: an in-app browser for researching sites and assessing material, a Chrome connection for existing browser workflows such as expense entry, and computer use for desktop applications such as Apple Notes.
GameStop CEO Ryan Cohen argues that his rejected roughly $56 billion bid for eBay would turn the marketplace into a larger, more profitable business by combining its platform with GameStop’s stores, gaming expertise and refurbished-technology operations. Cohen says the stores could serve as local hubs for authentication, fulfillment and live commerce, while $2 billion in first-year cost reductions would support the deal’s economics. He has offered broad assurances on financing and an investment-grade credit profile, but has not disclosed the proposed capital structure or said whether he will raise the bid.
Saronic Technologies plans to spend $3.2 billion building Port Alpha, a Brownsville shipyard for medium- and large-class autonomous surface vessels that CEO Dino Mavrookas says would address a severe shortfall in U.S. shipbuilding capacity. He says the greenfield project could create 10,000 jobs and eventually reach 2 million gross tons of annual capacity, while Governor Greg Abbott argues Texas can supply the workforce through its technical colleges and expanding Rio Grande Valley infrastructure. Mavrookas also casts the yard as a national-security asset following the reported combat use of Saronic technology in the Iran conflict.
Computer scientist Károly Zsolnai-Fehér argues that AI coding assistance can weaken learning when it substitutes for understanding rather than supporting it. In the small study he examines, mostly junior developers using an AI assistant finished a Python task only marginally faster—a difference that was not statistically significant—but scored 17 points lower on a subsequent quiz, with the largest weakness in debugging. His recommendation is to automate work already understood and use AI as a tutor for unfamiliar tasks.
OpenAI’s Romain Huet argues that developers should stop treating the next model release as a reason to postpone ambitious projects. With OpenAI shipping models roughly every six weeks, he says users quickly normalize new capabilities without fully testing what current tools can do; his advice is to use Codex on the “crazy ideas” and passion projects that have been left aside.
George Mack and Chris Williamson argue that many habits, norms and judgments people treat as natural are products of local circumstances, social pressure and unexamined defaults. From AI writing and procurement rules to male vulnerability and anxiety, they return to the same question: whether a reaction reflects considered judgment or inherited programming. Their practical case is for reflection that alters conduct rather than merely supplying another idea to admire.
Brent Barcimo, a 14-year employee at a major aerospace company, says Stanford Online’s Engineering Leadership Program was valuable because its lessons could be applied immediately in his work. He argues that the harder task was not completing the coursework but finding the confidence to use and share what he had learned. Barcimo also sees the Stanford credential as evidence of continued professional development and a meaningful addition to his résumé.
Former national security officials Stephen Hadley, Robert O’Brien and Michèle Flournoy, alongside AEI’s Marc Thiessen, disagreed over whether Trump’s use of military force, tariffs and allied pressure can produce arrangements that last beyond his presidency. Thiessen and O’Brien argued that coercion has forced overdue changes in allied defense spending and constrained adversaries; Flournoy and Hadley warned that military damage, strained alliances and inconsistent signals do not by themselves constitute strategy. Their central question was whether immediate leverage can be converted into durable deterrence, political settlements and institutions.
Grant Sanderson of 3Blue1Brown defines cross-entropy as the average coding cost of data generated by one probability distribution when encoded using another distribution’s assumed probabilities. In language-model pre-training, that cost is the average negative log probability assigned to observed next tokens, so minimizing it pushes the model toward the distribution underlying its data. In distillation, a smaller model minimizes cross-entropy against a larger model’s full probability distribution; KL divergence is the excess coding cost above the entropy of a matched code.
Democratic transitions in the Americas risk losing legitimacy when security-first or externally coordinated roadmaps lack credible evidence that power will change hands and domestic political ownership. Leopoldo López argued that Venezuela needs visible steps toward a credible election and the dismantling of repressive institutions; OAS secretary general Albert Ramdin backed country-specific pathways that extend beyond elections to durable governance. Former Haitian foreign minister Dominique Dupuy warned that Haiti’s prolonged interim rule shows how transition plans can perpetuate exclusion when the public remains outside the political process.
U.S. Trade Representative Jamieson Greer argues that U.S. trade policy should use tariffs, market access and security conditions to redirect production toward the United States and North America while limiting exposure to China-linked supply chains. He describes a durable hierarchy of access—not a return to pre-tariff trade—in which companies can plan around known rates but must meet stricter origin, investment and export-control rules. China, in his account, is a commercial relationship to be bounded rather than reformed through negotiation.
NVIDIA argues that Japan’s manufacturing disciplines, engineering culture and long relationship with robotics make it a natural setting for AI factories—computing systems that produce intelligence for scientific, engineering and industrial work. The company presents this as the next phase of a three-decade relationship that began in gaming and later extended into accelerated computing, framing AI infrastructure and robotics as the future it seeks to build with Japan. It identifies no specific AI-factory deployments, however, instead making its case through industrial fit and ambition.
Boston College biologist Thomas Seyfried argues that cancer begins primarily with chronic mitochondrial damage, which pushes cells toward glucose- and glutamine-dependent fermentation; in his account, many genetic mutations are downstream effects rather than the initiating cause. He says that model supports using nutritional ketosis and other metabolic measures alongside chemotherapy, radiation and immunotherapy to pressure tumours’ fuel supply. The source notes that this remains a contested minority view, and that such approaches are experimental rather than established cancer treatment.
Fatih Birol of the International Energy Agency argues that energy security is shifting from a just-in-time model built around cheap supply to a more expensive “just in case” system of reserves, alternate routes, grids and trusted partners. Alongside Meghan O’Sullivan of Harvard and former U.S. deputy energy secretary Daniel Poneman, he contends that disruptions from Hormuz to mineral refining expose a wider set of dependencies: electricity and nuclear fuel, transmission capacity, processing and long-term finance. The central constraint, they argue, is not simply resource availability but the ability to build and sustain resilient supply systems.
John Coogan argues that Stripe’s $53 billion offer for PayPal is a wager that its consumer accounts, bank-linked relationships, checkout presence and cash flow can be worth far more under stronger management. He and Jordi Hays say the transaction depends less on identifying those assets than on whether Stripe, alongside Advent International, can restructure and integrate a 25,000-person legacy fintech without eroding them. The show applies a similar test to OpenAI’s reported AI speaker and Chip Motors’ autonomous neighborhood EV: appealing concepts still have to clear the harder work of building, operating and delivering them.
OpenAI’s Danielle Zaghian presents scheduled tasks in ChatGPT as a way to turn recurring work into defined workflows: specify the sources, output, schedule and limits, then have ChatGPT run the task when needed. Her examples range from a read-only weekday brief that prioritizes work across Slack, email and calendar to a weekly feedback triage that ranks issues, assigns owners and drafts team updates. Converting the latter into a Workspace Agent moves it from a laptop-based task into a shared cloud workflow available in ChatGPT or Slack.
Cursor’s Lee Robinson argues that progress in AI-native software development depends less on any single training run than on a system that turns product failures into harder evaluations, targeted reward signals, and successive model updates. At Cursor, user feedback and agent usage feed an outer loop that identifies failures, while an inner loop builds the tasks, data, and training methods to address them. Stronger top-level models can then serve as judges and reward models for later runs, making the training machinery itself more capable.