2026-06-03 · Hosted by Caroline Hyde, Ed Ludlow · Bloomberg / iHeartMedia

Bloomberg Tech led with two landmark AI capital stories: Alphabet announcing an $80 billion equity raise — one of the largest equity deals in history — to fund AI infrastructure, and Anthropic confidentially filing an S-1 for its IPO, pulling ahead of rival OpenAI in the race to go public. Bloomberg Intelligence warned hyperscaler cumulative CapEx could approach $800 billion this year and surpass $1 trillion next year, with projections of $5 trillion over five years. The episode also covered Impulse Space closing a $500 million round at a $4 billion valuation, Nvidia CEO Jensen Wong calling Marvell the next trillion-dollar company (sending shares +~30%), Perplexity announcing tripled revenue to ~$500 million ARR, HPE raising its 2026 and issuing 2027 guidance early on the back of massive AI server demand, and the White House releasing a new AI executive order with a voluntary cybersecurity vetting framework.

The AI infrastructure buildout has consumed virtually all available debt capacity for the largest technology companies, forcing them into equity markets for the first time in decades. Alphabet’s $80 billion raise — following $85 billion it had already raised in credit markets — signals that CapEx spending trajectories are even more aggressive than analysts modeled, with Bloomberg Intelligence forecasting the industry approaching $800 billion in cumulative spend this year and over $1 trillion next year. This shift from debt to equity financing marks a structural change in how the AI buildout is capitalized.

Three of the largest AI companies in the world — SpaceX (pricing June 11), Anthropic (confidential S-1), and OpenAI (likely September) — are racing to the public markets simultaneously. PitchBook analysis shows combined proceeds would exceed all US VC-backed IPO proceeds from the past decade, entering genuinely uncharted territory. The competitive dynamics are circular: these companies rely on each other for compute, share investors on cap tables, and are all pitching the same transformative technology thesis to the same pool of institutional capital.

The current earnings cycle is revealing that AI spending is no longer exclusively a hyperscaler story. HPE’s triple-digit server demand growth, Marvell’s soaring optical networking business, and SK Hynix’s capacity commitments show the AI build is cascading down the full hardware value chain — from GPU clusters to networking gear, memory, and space infrastructure. Jensen Wong’s Marvell endorsement crystallized the market’s recognition that custom silicon and optical connectivity are becoming the next bottlenecks after GPUs.

HPE CEO Antonio Neri and Goldman Sachs CEO David Solomon both described a structural shift where enterprise companies are moving from experimentation to production AI deployments. The shift from cloud-only to hybrid/on-premise architectures — driven by governance, data privacy, and cost concerns — is creating a new wave of enterprise infrastructure spending. HPE’s own 250 production AI use cases out of 1,200 tested exemplify where large enterprises are heading.

Arm CEO René Haas and SK Hynix’s leadership both warned that the AI compute shortage could persist well beyond 2030. The traditional boom-bust cycle for memory is being challenged by an unprecedented, prolonged demand signal, but chipmakers remain cautious about over-building given their experiences in 2022. The emerging consensus is that hyperscalers may need to co-invest in fab capacity to share risk with manufacturers — a model Oracle already pioneered earlier in 2026.

The White House AI executive order — watered down from 90-day to 30-day review periods after industry pushback — reflects the ongoing tension between national security concerns and tech sector growth. The voluntary nature of the framework is a win for AI labs’ product timelines. However, the simultaneous release of Anthropic’s Mythos to 150+ global organizations signals that AI capabilities are outpacing any regulatory framework’s ability to manage them. —-

The episode captured a market that is broadly bullish on AI infrastructure but beginning to show signs of reflexivity, with analysts noting circular dependencies among AI IPO candidates and growing equity supply concerns.

Dilution from historic equity issuance: Alphabet’s $80B raise — combined with $85B already raised in credit — represents over $165B in new capital this year; stock buybacks will decline materially

Circular IPO dependencies: SpaceX, Anthropic, and OpenAI share overlapping investors and compute relationships; a SpaceX IPO flop could delay Anthropic’s and OpenAI’s timelines

Capital absorption limits: PitchBook identified that combined AI IPO proceeds would exceed the entire past decade of US VC-backed listings; market capacity to absorb is untested

Memory shortage and pricing pressure: Rising DRAM/NAND costs are flowing through to HPE and other hardware vendors; Arm CEO cautioned that memory demand could remain constrained through 2030

ROI validation still pending: Bloomberg Intelligence and multiple analysts noted that confidence in AI monetization is rising but has not yet been fully proven at scale across enterprises

Regulatory timing risk for IPOs: Confidential filings are not guarantees; Anthropic’s IPO still depends on SpaceX success and market conditions; regulatory review of frontier models adds unpredictability

AI copyright litigation: Perplexity facing CNN lawsuit; outcome could affect business model for AI aggregators broadly

Buyback reduction: Alphabet explicitly signals reduced buyback activity as CapEx absorbs capital, a headwind for shareholder return metrics

This episode was covered in today’s The Market Signal — 2026-06-03, a cross-source synthesis of multiple podcast reports.