Trump's First 100 Days, Tariffs Impact Trade, AI Agents, Amazon Backs Down

Chamath and Sacks give Trump roughly a B+/A- for the first 100 days, citing an A+ on sealing the border, attracting ~$1T in foreign direct investment, and rapidly dismantling DEI/woke frameworks in government and culture. They argue Trump has begun ‘reprivatizing’ the economy via deregulation, cutting federal headcount, ending Biden’s AI and EV mandates, and unleashing energy. However, failures on promised document releases (Epstein, MLK, full JFK) and inconsistent, confusing communication around tariffs and policy changes drag the grade down.

Ryan Petersen reports a roughly 60% drop in ocean freight bookings from China to the US after Trump’s reciprocal tariffs escalated from an initially signaled 54% to roughly 154% on Chinese imports, far exceeding expectations. Because duties apply based on departure date (post–April 9) and must be paid at or shortly after arrival, importers are scrambling to defer payments via bonded warehouses in the US, Mexico, and Canada, betting that tariff rates will be negotiated down. The shock timing and scale are already prompting layoffs, panic among small brands, and aggressive inventory pull-forwards.

Chamath and Sacks argue tariffs are exposing dangerous US dependencies on China for rare earths, batteries, pharma APIs, and critical manufacturing capacity—dependencies that could cripple US options in a China–Taiwan or China–India conflict. They see Trump’s move as breaking a 25-year ‘globalist’ consensus that let China use subsidies and WTO rules to hollow out US capability. Aaron and Ryan agree on the national-security diagnosis but attack the method: policy has been rolled out via ‘shock and awe’ barrel rolls, creating chaos for small businesses and Fortune 500 supply chains that could have been steered using clearer carrots, better planning, and more predictable timelines.

Tariffs have thrown a spotlight on Amazon’s marketplace dynamics. Ryan notes about 60% of Amazon sellers are foreign, often Chinese entities with no US legal presence, making it easy to misdeclare valuations, classifications, or safety standards and hard for regulators to enforce rules. Many US-based Amazon sellers feel ‘abandoned’ as Amazon facilitates foreign competitors undercutting them on price. While Trump publicly pressured Amazon over how it surfaced tariff costs to consumers, the panel sees a deeper opportunity: reform de minimis/import rules, clamp down on foreign seller abuse, and potentially steer consumers toward more domestic products—if the administration can avoid case-by-case ‘whack-a-mole’ and instead address root causes.

Sacks and Levie describe AI agents as long-running, multi-step systems that can operate across many SaaS apps—searching, browsing, editing documents, running code terminals, and executing workflows. Standards like MCP will let agents plug into dozens of enterprise tools. Levie argues this transforms software economics: instead of charging per user-seat to marginally boost productivity, vendors can charge for AI ‘workers’ that perform entire workflows, tapping into labor budgets instead of only software budgets. Many early use cases are not replacing existing work but enabling tasks that were previously uneconomical—e.g., Flexport’s AI calling thousands of truck drivers daily or enterprises finally reviewing long-ignored contracts and invoices.