Bloomberg Tech – Episode Summary
Episode Title: Amazon Shares Soar on Cloud Growth
Date: October 31, 2025
Host: Caroline Hyde (New York) & Ed Ludlow (San Francisco), Bloomberg
Guests:
- Brad Ericsson (RBC Capital Markets)
- Carolina Milanesi (Creative Strategies)
- Jeff Lawson (Twilio CEO)
- Gen Wang (Reddit CEO)
- Michael Intrater (Core Scientific CEO)
- Cameron Adams (Canva Co-Founder)
- Alexandra Levine (Bloomberg Social Media Reporter)
- Amy Webb (Future Today Strategy Group)
- Rachel Metz (Bloomberg Startup Reporter)
Episode Overview
This episode delivers a fast-paced breakdown of major tech earnings and broader industry pivots, focusing on Amazon’s stock surge driven by a rebound in its cloud business. The conversation expands to Apple’s shifting fortunes, Twilio’s AI-fueled comeback, Reddit’s ad and data strategy, and dramatic developments for Nvidia, Core Scientific, and CoreWeave amidst the AI infrastructure boom. The show also examines the latest on TikTok in the U.S.-China regulatory crossfire, trends in AI startup funding, and platform-level risks like copyright and localization.
Key Segments & Insights
1. Amazon’s Cloud Comeback Drives Record Highs
[03:16–06:58]
-
Brad Ericsson (RBC): Amazon posted a nearly 11% uptick, hitting a record high thanks to a “reacceleration to 20%” growth in the U.S. cloud (AWS) business.
-
Key drivers:
- Speedy capacity expansion ("getting this capacity deployed faster than we would have thought… gave some nice detail around some of the power, the gigawatts they’ve been able to deploy" – [03:35])
- Vertical integration, especially with in-house Trainium AI chips, fueling "material" revenues for next year, largely via Anthropic.
-
Notable Quote:
“Training them to go with your second part first. Training too, is largely being used for Anthropic... going to double the number of chips at the Rainier site by the end of the year. So it’s… going to continue to scale up and drive a ton of revenue growth.” — Brad Ericsson [04:21] -
Power infrastructure is a key challenge: “The power grid is not set up for these data centers and they have massive needs... it's massively important [to go vertical].” — Ericsson [04:40]
-
Generative AI’s Reach:
- Tools like “Rufus” could add $10B to Amazon: “Rufus is really intended to… add a tailwind to your shopping experience… literally gets you to buy more stuff.” — Ericsson [05:54]
- Corporate job cuts (14,000 this week) likely linked to AI-led efficiency, even if officially described as flattening management.
- "Clearly generative AI... is very likely that it is contributing to job losses right now." — Ericsson [05:54]
2. Apple: Innovation, China Headwinds & Services Growth
[07:43–11:20]
- Carolina Milanesi (Creative Strategies):
- iPhone sales beat expectations (~8% FY Q4 growth), but China sales lagged due to economic uncertainty and supply chain shift.
- “A new model every September brings people back in stores and upgrade to previous models as well.” — Milanesi [07:43]
- China’s Q1 (Chinese New Year) will be the key "big quarter" as supply constraints ease.
- Services & Content:
- Apple’s services arm (e.g., F1 deal, music) is the current “reliable” growth engine, with Gen Z and Millennials being focal demographic.
- “The opportunity for Apple... is the engagement side and that also over time will come into play with Apple Vision Pro.” — Milanesi [10:15]
- AI Outlook:
- Siri upgrades promised for 2026 as Apple pivots to more personal and context-aware AI assistance.
3. Twilio: Voice AI & Platform Expansion
[13:56–18:15]
- Jeff Lawson (CEO):
- Q3 marked by broad product strength, especially in “voice AI” and messaging.
- Key enterprise customers (Lyft, Reddit, Dell, Uber, Shopify) are driving usage, though AI startups adopt faster.
- “If you have contextual data, which is a big part of our story… that data is what really allows the customer's problem to get solved.” — Lawson [15:37]
- Small but strategic acquisition of Stitch broadens Twilio’s platform, especially for agent/identity solutions.
- On margin pressure: growth in messaging and carrier fees “doesn’t impact our ability to generate gross profit and… has not impeded our ability to drive operating leverage and free cash flow.” — Lawson [17:24]
4. Reddit: Ads, Data Licensing, and AI Localization
[18:36–22:42]
- Gen Wang (CEO):
- Active advertisers up 75% YoY in Q3; robust ad marketplace diversification across geos and verticals.
- “Our traffic today is about 55% outside of the US… every country that is a non-English country is just an opportunity to build another Reddit.” — Wang [19:36]
- AI used for both machine translation and as a sales/advertising tailwind.
- Data licensing deals (OpenAI, Google) are a valuable secondary business, but “core business is advertising… where we’re focused.” — Wang [21:04]
- Chatbot/LLM relationships "under heavy construction" and rapidly evolving: “...our data is highly appreciated, highly used and cited… topmost source of domain in Q3.” — Wang [21:55]
5. Nvidia: China Hopes, Korea Push, and Sovereign AI
[22:42–25:00]
- Jensen Huang (CEO) via reporting:
- Nvidia just hit $5 trillion market cap; aims to sell Blackwell chips in China “someday.”
- “Korea has a chance to be one of the world’s major AI hubs... the journey for Korea, this is a perfect example.” — Huang, via Brad Ericsson [23:58]
- Nvidia also spurred the so-called “fried chicken effect” in South Korea as stocks surged merely from a restaurant photo-op.
- Bloomberg’s Analysis:
- Nvidia’s international push for sovereign AI projects, especially in South Korea, positions it as a global infrastructural player.
6. Core Scientific vs. CoreWeave: Data Center Arms Race
[27:17–33:17]
- Michael Intrater (Core Scientific CEO):
- Core Scientific’s failed takeover by CoreWeave was purely a matter of price; business relationship remains stable.
- “Our footprint within the Core Scientific ecosystem is about 580 megawatts... This quarter we’ve signed over 600 megawatts of… data center infrastructure outside and... outside of Core Scientific.” — Intrater [28:55]
- Acquisitions are either strategic (companies like Marimo, Weights & Biases) or opportunistic (infrastructure providers).
- Capital needs remain high but justified by accelerating demand: “Our growth continues to rage along. I mean it’s just amazing how fast we’re growing...” — Intrater [30:56]
- Meta’s heavy infrastructure investment is seen as positive for the space, not concerning.
7. Canva: Copyright, AI, and Partnerships
[33:40–35:47]
- Cameron Adams (Co-Founder):
- Canva stresses full transparency and indemnity against copyright risks for AI-generated assets (“Canva Shield”).
- “Every interaction you have with AI can be controlled in Canva. We give you all the switches and levers…” — Adams [33:40]
- Emphasizing flexibility in partnering with top AI model providers, in-house R&D, and fostering a large third-party developer ecosystem.
- Over 100 proprietary models now run inside Canva’s suite.
8. TikTok in U.S.-China Limbo & AI Cold War
[38:30–45:03]
- Alexandra Levine (Bloomberg) & Amy Webb (Future Today):
- Despite executive order and U.S.-China talks, no clarity on TikTok’s fate; skepticism persists about “true” separation from ByteDance.
- “There really has to be no operational relationship… several aspects of the deal show... ByteDance will… still have oversight…” — Levine [39:19]
- Amy Webb: Relationships have “de-escalated but didn’t end the AI Cold War.”
- U.S.-China rivalry is now about “transistors not tariffs.”
- Nvidia’s China prospects look steadier, but the real long-term play is China’s aim to build local chip supply chains and dominate industrial modernization.
- “China’s top priority is building… a modern industrial system… heavily investing in frontier sectors… all of this requires those advanced chips.” — Webb [42:40]
- On rare earths: Western self-reliance is likely "years away at best."
9. AI Startup Landscape & Globalization
[45:03–47:44]
- Rachel Metz (Bloomberg):
- 2025 has seen nearly $200B in VC inflows to AI startups, with emphasis on international diversity and specialized application verticals.
- “Lovable in Sweden…has grown tremendously quickly… really trying to democratize the idea of building something on the Internet…” — Metz [47:16]
- Cross-category innovation: infrastructure (Mistral in France), coding democratization (Lovable, Replit), and creative (Black Forest Labs in Germany).
Notable Quotes & Timestamps
- “Reacceleration to 20% on us in Q3… really got the stock going.” — Brad Ericsson [03:16]
- “The power grid is not set up for these data centers and they have massive needs.” — Brad Ericsson [04:40]
- “People still want to rely on the device that is in their pocket 24/7 but make it look different.” — Carolina Milanesi [07:43]
- “You don’t want just cool tech. You want problems to get solved.” — Jeff Lawson [15:37]
- “Every country that is a non-English country is just an opportunity to build another Reddit.” — Gen Wang [19:36]
- “Korea has a chance to be one of the world’s major AI hubs.” — Jensen Huang (via Brad Ericsson) [23:58]
- “Every interaction you have with AI can be controlled in Canva. We give you all the switches and levers…” — Cameron Adams [33:40]
- “There really has to be no operational relationship [between TikTok and ByteDance]… but there are a lot of points that make it seem like there is in fact going to be an operational relationship.” — Alexandra Levine [39:19]
- “China’s top priority is building what it’s calling a modern industrial system... all of this requires those advanced chips.” — Amy Webb [42:40]
- “They really are trying to democratize the idea of building something on the Internet… Lovable in particular is focused on helping you build, say, a website, even if you don’t know anything about coding.” — Rachel Metz [47:16]
Further Highlights & Industry Tone
- The episode maintained an energetic, analytical, and lightly skeptical tone—balancing excitement about rapid tech growth and innovation with questions about sustainability, regulatory risk, and global supply chain fragility.
- There’s broad consensus that AI—especially generative and contextual, voice-enabled, or localized—remains the driver of both growth and disruption across hardware (Apple), cloud (Amazon), SaaS (Twilio), and social (Reddit).
- Infrastructure (power/data/capital) is now a central battleground, with Readiness, regulatory clarity, and vertical integration distinguishing winners.
- U.S.-China tech rivalry is nuanced: optimism on some fronts, but persistent core dependencies (chips, rare earths, regulatory friction).
Episode Structure Guide
- [03:16–06:58] Amazon AWS, AI, and Job Cuts
- [07:43–11:20] Apple Earnings, China, and Services
- [13:56–18:15] Twilio’s Q3 & Voice AI Push
- [18:36–22:42] Reddit’s Ad Model & Globalization
- [22:42–25:00] Nvidia’s Global Play
- [27:17–33:17] Core Scientific / CoreWeave Drama
- [33:40–35:47] Canva: Copyright & AI
- [38:30–45:03] TikTok, U.S.-China Tech Tensions
- [45:03–47:44] AI Startup Global Trends
This summary captures the dynamism, critical analysis, and major takeaways from the October 31, 2025 episode of Bloomberg Tech, offering a structured guide for listeners and non-listeners alike.
