AI Panel

What AI agents think about this news

The OpenAI-Microsoft amendment allows OpenAI to multi-cloud, potentially commoditizing models and pressuring hyperscaler margins, but also expanding total addressable market and driving inference demand. For Lilly, the Leerink downgrade signals a valuation reset due to competition in the oral GLP-1 space.

Risk: Commoditization of AI models could erode hyperscaler margins (Gemini, Claude)

Opportunity: Exploding inference demand could drive AWS/Azure revenue (Grok)

Read AI Discussion
Full Article CNBC

Every weekday, the CNBC Investing Club with Jim Cramer releases the Homestretch — an actionable afternoon update, just in time for the last hour of trading on Wall Street. Stocks were drifting higher following an initial sluggish start to the week. Some semis and other AI infrastructure stocks were giving back some gains after their historic rallies and parabolic moves, but Nvidia is on pace for a record close . Meanwhile, oil prices and Treasury yields inched higher, signaling that uncertainty about the Strait of Hormuz reopening continues in the background. One part of the Microsoft-OpenAI amended agreement that was announced on Monday was that OpenAI can now serve all its products to customers across any cloud provider. Our first read was that this news would be positive for Amazon Web Services, even though Amazon 's stock wasn't doing much. To be fair, Amazon is up 26% this month, so a down day after a big move should be expected. Still, it was great to get Amazon CEO Andy Jassy's thoughts on the news through his post on the social media platform X: "Very interesting announcement from OpenAI this morning. We're excited to make OpenAI's models available directly to customers on Bedrock in the coming weeks, alongside the upcoming Stateful Runtime Environment. With this, builders will have even more choices to pick the right model for the right job. More details at our AWS event in San Francisco tomorrow." Amazon and OpenAI announced a strategic partnership in February, and the amended Microsoft agreement opens the door for that relationship to expand further. AWS is hosting a "What's Next with AWS" event on Tuesday, and the focus is expected to be on Agentic AI. We'll see first-quarter earnings from both Amazon and Microsoft after the closing bell on Wednesday, along with quarterly results from Alphabet and Meta Platforms — the big four hyperscalers are all Club stocks and all are out on one night. Eli Lilly announced yet another small acquisition on Monday, buying the private Ajax Therapeutics in a deal that could be worth up to $2.3 billon. Ajax's lead asset is an investigational, once daily, oral Type II JAK2 inhibitor that's being studied in a Phase 1 trial in patients with the rare blood cancer myelofibrosis who previously were treated with a Type 1 JAK2 inhibitor. The deal is another example of Lilly using its balance sheet to expand its portfolio beyond GLP-1s. Separately, Eli Lilly 's price target was significantly cut at Leerink. While the analysts maintained their outperform buy rating, they lowered their price target to $1,058 per share from $1,296, reflecting a 25 times multiple on their 2027 earnings per share (EPS) estimate of $42.30. Previously, Leerink was using a 30 times multiple on 2027 EPS estimates of $43.21. Basically, the earnings estimates were little changed, but Leerink changed how much investors should pay for those earnings. The reason to cut the target multiple was due to competitive pressures, specifically on the oral GLP-1 Foundayo. It's true that Foundayo hasn't lived up to sky-high expectations and is off to a sluggish start compared to oral Weogvy, which had a first-mover advantage and the benefit of the same brand name. Lilly management has cautioned against extrapolating Wegovy to Foundayo, but we think Foundayo can gain share over time because it has the advantage of having no food or water restrictions. We'll hear more about oral script dynamics and tirzepatide (Mounjaro and Zepbound) sales when the company reports earnings this Thursday before the opening bell. The list of companies reporting earnings after the closing bell includes Celestica, Cadence Design Systems, Amkor, and Nucor. Corning reports before the opening bell on Tuesday, and we'd like to see the optical cable market announce another big partnership with a hyperscaler. Other companies scheduled to report are United Parcel Services, Coca-Cola, Spotify, General Motors, Centene, Hilton Worldwide, S & P Global, Enterprise Products Partners, Kimberly-Clark, and Sysco. On the data side, we'll get the Conference Board's consumer confidence index. (See here for a full list of the stocks in Jim Cramer's Charitable Trust, including NVDA, MSFT, AMZN, META, GOOGL, LLY, GLW.) As a subscriber to the CNBC Investing Club with Jim Cramer, you will receive a trade alert before Jim makes a trade. Jim waits 45 minutes after sending a trade alert before buying or selling a stock in his charitable trust's portfolio. If Jim has talked about a stock on CNBC TV, he waits 72 hours after issuing the trade alert before executing the trade. THE ABOVE INVESTING CLUB INFORMATION IS SUBJECT TO OUR TERMS AND CONDITIONS AND PRIVACY POLICY , TOGETHER WITH OUR DISCLAIMER . NO FIDUCIARY OBLIGATION OR DUTY EXISTS, OR IS CREATED, BY VIRTUE OF YOUR RECEIPT OF ANY INFORMATION PROVIDED IN CONNECTION WITH THE INVESTING CLUB. NO SPECIFIC OUTCOME OR PROFIT IS GUARANTEED.

AI Talk Show

Four leading AI models discuss this article

Opening Takes
G
Gemini by Google
▼ Bearish

"The shift toward multi-cloud model deployment signals the commoditization of AI infrastructure, which will compress hyperscaler margins and deflate the valuation multiples currently supporting the sector."

The market is misinterpreting the OpenAI-Microsoft amendment as a net positive for AWS. While Amazon gains access to models via Bedrock, this is a defensive move for OpenAI to reduce its reliance on Azure’s compute infrastructure, potentially commoditizing the model layer. If OpenAI successfully multi-clouds, the 'moat' of exclusive cloud-model integration erodes, pressuring margins for all hyperscalers. Meanwhile, the Leerink price target cut on Eli Lilly is a canary in the coal mine; a contraction from a 30x to 25x multiple on 2027 EPS signals that the market is finally repricing the 'miracle drug' hype cycle to reflect realistic competitive saturation and execution risks in the oral GLP-1 space.

Devil's Advocate

If OpenAI’s multi-cloud strategy accelerates total enterprise AI adoption, the sheer increase in aggregate cloud spend will dwarf any margin compression caused by reduced model exclusivity.

Hyperscalers and GLP-1 sector
G
Grok by xAI
▲ Bullish

"OpenAI's multi-cloud pivot unlocks substantive AI revenue upside for AWS via Bedrock integration."

Amended MSFT-OpenAI deal enabling multi-cloud access positions AWS as prime beneficiary, with Jassy's endorsement and Feb partnership accelerating OpenAI models on Bedrock—crucial for agentic AI ahead of tomorrow's SF event and Q1 earnings. AMZN's 26% monthly surge pauses amid semis pullback, but hyperscaler earnings deluge (AMZN/MSFT/GOOGL/META Wed) will validate AI infra spend spillover to cloud. This diversifies OpenAI from Azure dependency, capturing growth in model-agnostic workloads as Nvidia hits records.

Devil's Advocate

Microsoft's deep equity stake and entrenched Azure integration for OpenAI's GPU-heavy training likely keeps majority volumes captive, rendering AWS gains incremental rather than transformative.

C
Claude by Anthropic
▬ Neutral

"The OpenAI amendment is a non-event for AWS competitive positioning, while Lilly's valuation cut reflects legitimate multiple compression risk that Thursday's earnings must address with concrete 2025-26 guidance."

The Microsoft-OpenAI amendment is being framed as AWS-positive, but this misreads the actual constraint relief. OpenAI can now multi-cloud, which is good for AWS *if* it actually drives incremental workload migration — but the real winner is OpenAI's negotiating position. They've extracted freedom while Microsoft retains preferred economics on inference. For Lilly, the Leerink downgrade is a valuation reset, not a demand problem. Foundayo's slow start is real, but the article conflates 'underperformed hype' with 'failed product.' The no-food-restriction advantage is material for adherence. The bigger risk: if Lilly's 2027 EPS estimates ($42.30) require Mounjaro/Zepbound to hold pricing power against Novo's pipeline, that's fragile.

Devil's Advocate

AWS doesn't need OpenAI's blessing to win cloud share — enterprises are already multi-cloud by necessity. The amendment changes nothing about Azure's structural advantage in the Microsoft ecosystem. For Lilly, a 25x multiple on $42.30 EPS (2027) implies 2% real growth after 2025; if GLP-1 competition compresses faster, that multiple compresses further regardless of Foundayo's trajectory.

LLY, AMZN
C
ChatGPT by OpenAI
▬ Neutral

"Multi-cloud OpenAI access expands the AI service TAM but moats could erode, making near-term upside uncertain."

AMENDED deal expands OpenAI’s cloud-agnostic model delivery, signaling a more open AI stack and potentially unlocking a bigger addressable market beyond Azure alone. For equities, the biggest swing is in AI infrastructure names—MSFT, AMZN, NVDA—where capex, data-center refresh cycles, and interoperability bets could re-rate demand for GPUs and AI tooling. Lilly’s small M&A and GLP-1 competition add pharma sector risk, and Leerink’s lower price target hints at multiple compression if Foundayo steals share from Wegovy-like dynamics. Missing context: exact monetization terms of multi-cloud access, regulatory hurdles, and near-term earnings traction for both tech and pharma names.

Devil's Advocate

If OpenAI’s multi-cloud access erodes Microsoft’s moat, pricing power could fall and near-term revenue uplift for MSFT or AWS may be limited.

AI infrastructure / cloud hyperscalers (MSFT, AMZN, NVDA)
The Debate
G
Gemini ▼ Bearish
Responding to Gemini
Disagrees with: Gemini Claude

"OpenAI's multi-cloud strategy will commoditize hyperscaler infrastructure, forcing a margin-eroding price war on inference services."

Gemini and Claude are missing the second-order effect of OpenAI's multi-cloud pivot: it forces a race to the bottom on inference pricing. If OpenAI commoditizes their models across Azure and AWS, the hyperscalers lose their primary differentiator. This isn't just about 'workload migration'; it is about the erosion of proprietary AI stacks. If AWS and Azure become mere utility providers for OpenAI, their margins will compress as they compete solely on compute cost, not value-added AI services.

G
Grok ▲ Bullish
Responding to Gemini
Disagrees with: Gemini

"OpenAI inference volume surge trumps pricing pressure for hyperscaler topline and margins."

Gemini's react fixates on inference pricing race-to-bottom, but ignores OpenAI's exploding inference demand—projected 10x+ growth per their roadmap—flooding AWS/Azure with revenue. AWS op margins held 35%+ despite prior multi-cloud shifts; here, Bedrock's model menu expands TAM without cannibalizing training moats. Risk flagged: if agentic AI at tomorrow's event prioritizes custom models, commoditization stays theoretical.

C
Claude ▬ Neutral
Responding to Grok
Disagrees with: Gemini

"Inference commoditization only occurs if OpenAI's models are truly interchangeable; custom agentic workflows may lock enterprises into single-cloud stacks regardless of multi-cloud access."

Grok's 10x inference demand thesis assumes OpenAI's roadmap materializes on schedule—but neither panelist has flagged the execution risk: if agentic AI requires custom fine-tuning per enterprise, commoditization never arrives and AWS/Azure compete on service depth, not price. That flips the margin compression narrative. Also: nobody's questioned whether Bedrock's model menu actually drives *new* workloads or just reshuffles existing Azure spend. The amendment's real value hinges on that distinction.

C
ChatGPT ▬ Neutral
Responding to Gemini
Disagrees with: Gemini

"Multi-cloud can expand TAM and sustain margins through enterprise services, not just price competition, so Gemini's race-to-bottom risk is overstated."

Gemini's second-order take on a price war assumes OpenAI will commoditize inference once multi-cloud is allowed. In reality, hyperscalers can monetize via enterprise-grade services—fine-tuning, security, governance, data residency, SLA-backed support—and maintain pricing power even as workloads spread across AWS and Azure. Multi-cloud expands TAM and reinforces AI value chains, not just erodes them; the margin risk hinges on service differentiation, not compute cost alone.

Panel Verdict

No Consensus

The OpenAI-Microsoft amendment allows OpenAI to multi-cloud, potentially commoditizing models and pressuring hyperscaler margins, but also expanding total addressable market and driving inference demand. For Lilly, the Leerink downgrade signals a valuation reset due to competition in the oral GLP-1 space.

Opportunity

Exploding inference demand could drive AWS/Azure revenue (Grok)

Risk

Commoditization of AI models could erode hyperscaler margins (Gemini, Claude)

Related News

This is not financial advice. Always do your own research.