Originally published on CoreProse KB-incidents
Wall Street is shifting its AI focus.
Beyond GPU and data‑center winners, investors now want sub‑$15 names that can turn the AI buildout into recurring, usage‑linked cash flows.
AI‑native cybersecurity platforms like SentinelOne sit at the junction of two macro forces: rapidly advancing AI capabilities and the unavoidable need to secure the agents, data, and infrastructure that power them. Morgan Stanley sees nearly $3 trillion of AI‑related infrastructure investment by 2028, with over 80% still ahead and AI already driving about a quarter of U.S. GDP growth this year.[5] In that cycle, “security for AI” is part of the core monetization stack, not a side theme.
1. Macro AI Setup: Why Wall Street Wants More Than Just Chip Winners
Morgan Stanley argues markets are underestimating a non‑linear jump in large language model capabilities expected by mid‑2026.[3][7] Leading AI labs are guiding investors to expect self‑improving models, with scaling laws suggesting roughly 10x more compute can double a model’s “intelligence.”[7][8]
📊 Key macro shift
AI models are compounding in capability
Training compute at top U.S. labs is rising by order‑of‑magnitude steps[7][8]
Progress is starting to look exponential, not incremental[3]
This leap is funded by industrial‑scale capex. Morgan Stanley projects about $2.9 trillion of global data‑center construction through 2028, driven by compute demand far above current supply.[5] Over 80% of that spend is still ahead, so the AI stack is in mid‑build, not late cycle.[5]
💡 Why this matters for security
2024–2025: “Buy the shovels” (chips, data centers)
By 2026: shift to the Inference Era, where returns move to software and SaaS that turn capex into workflows and productivity[11]
Morgan Stanley also notes a mispricing: software and data firms have been punished on fears of AI disruption, even as profits hold up and proprietary data moats look durable.[4] That opens room for software names whose products are:
Directly tied to AI deployment
Hard to rip out once embedded
Monetized via recurring or usage‑linked models
AI‑driven cybersecurity fits this pattern. It monetizes the risk side of AI and scales with AI‑enabled workloads.
⚡ Section takeaway: As AI becomes a macro force and markets rotate from infrastructure to ROI, investors will seek software platforms that ride the capex wave. AI‑native cybersecurity is one of the clearest of those layers.
2. Why AI Security and Cyber Platforms Sit in the Sweet Spot
The AI stack now includes full ecosystems: dev environments, evaluation frameworks, security tooling, deployment pipelines, and governance layers.[1] Securing AI agents against prompt injection, data leakage, and unsafe behavior is becoming foundational.
OpenAI’s acquisition of Promptfoo—used by more than 25% of the Fortune 500 for red‑teaming and evaluation—is a clear signal.[10] Evaluation and security are shifting from “nice‑to‑have” tools to core infrastructure that will be deeply integrated into model platforms.[1][10]
💼 The emerging AI security stack includes:
Red‑teaming and adversarial testing
Continuous evaluation for safety and compliance
Monitoring agents for jailbreaks, data exfiltration, and policy violations[1][10]
Inside Security Operations Centers, AI is present but often not operationalized. The 2025 SANS SOC survey found:[2]
40% of SOCs use AI/ML tools without making them a defined part of operations
42% rely on out‑of‑the‑box models with no customization
When AI is deliberately integrated into SOC workflows, SANS finds it can improve:[2]
Capability and maturity
Process repeatability
Staff capacity and analyst satisfaction
These benefits align with autonomous, AI‑first endpoint and threat‑detection architectures that:
Encode detection logic
Learn from new signals at machine speed
Push updates across fleets automatically
As LLMs evolve from chatbots into code‑writing, system‑controlling agents, evaluation and monitoring become mission‑critical.[1] AI agents will increasingly:
Call internal APIs
Touch production data
Trigger financial and operational actions[1]
⚠️ Compromising the agent surface will resemble compromising a privileged human operator.
This creates a structural tailwind for AI‑native security vendors that can continuously test, learn, and defend in real time. SentinelOne’s positioning as an autonomous, AI‑driven platform—spanning endpoints, cloud, and identity—matches where enterprise AI is heading: agent‑heavy, workflow‑embedded, and highly security‑sensitive.
⚡ Section takeaway: As enterprises move from LLM experiments to production AI agents, security and governance become indispensable. Platforms that treat AI as the engine, not a bolt‑on, are best placed.
3. Positioning an AI Cybersecurity Name Like SentinelOne in the 2026 AI Cycle
SentinelOne’s strategic fit is clearer when the macro buildout is tied to workloads. OpenAI has reached an estimated $20 billion revenue run rate and has tripled compute capacity in each of the last two years to about 1.9 GW in 2025, with plans that could push capacity toward 6 GW in 2026.[6]
Each watt of expansion enlarges the attack surface:
More models and APIs
More data paths
More privileged machine identities
Meanwhile, Morgan Stanley’s “Intelligence Factory” model projects a U.S. power shortfall of 9–18 GW through 2028, or 12%–25% of required AI power.[7][8] Developers are responding by:
Converting Bitcoin mines into dense AI data centers
Deploying natural‑gas turbines
📊 Implication: Infrastructure growth is constrained and capital‑intensive, pushing investors toward assets that:
Enable the buildout efficiently (e.g., neocloud providers)
Or protect the higher‑value software and data layers on top[5][6]
Research on AI stocks for the 2026 “ROI Era” argues that software and SaaS firms that convert AI capex into sticky, monetized products should outperform pure hardware, especially in mission‑critical areas like security and compliance.[11]
SentinelOne, as an AI‑first cybersecurity platform, checks several boxes investors seek in under‑$15 AI names:
Direct linkage to AI adoption
- More AI agents and cloud workloads → more endpoints, identities, and APIs to defend
Usage‑aligned monetization
- Seat, workload, and data‑volume pricing can scale with AI deployment
Workflow embedding
- Once wired into SOC workflows and response playbooks, switching costs rise
💼 In portfolios, this places SentinelOne in the “application and security layer” bucket—an indirect but powerful way to play AI without paying GPU or data‑center multiples, and with upside from potential re‑rating if markets recognize AI security as structural, not cyclical.
⚡ Section takeaway: Power and infrastructure constraints heighten the value of software that makes each unit of AI capex safer and more productive. An AI‑native cybersecurity mid‑cap can be a leveraged yet more diversified way to express that theme.
4. Risk, Timing, and Portfolio Role for an Under‑$15 AI Security Stock
The forces that make AI security attractive also make it volatile. Morgan Stanley warns markets are not prepared for the speed of AI capability gains, implying repeated repricing of winners and losers.[3][7] AI‑labeled names, including security stocks, can overshoot both up and down as sentiment swings between “bubble” and “supercycle.”
Yet their analysts also note that AI disruption fears have often outrun fundamentals. Many “at risk” software and data firms still post solid earnings, and proprietary‑data businesses remain hard to replicate.[4] That can create opportunities when selling is indiscriminate.
📊 Nasdaq commentary on AI investing mistakes highlights three pitfalls relevant to a sub‑$15 security name:[12]
Over‑concentrating in a few high‑beta AI stocks
Betting only on one theme (chips, data centers, etc.)
Treating AI as a short‑term trade instead of a multi‑year thesis
The OpenAI–CoreWeave relationship shows both upside and risk. CoreWeave, a leading neocloud provider, has surged as it rents GPU capacity to OpenAI, but its stock has been extremely volatile, at one point rising over 300% then dropping more than 60% as bubble fears grew.[6] That sensitivity underscores why investors may want to balance infrastructure bets with application‑layer plays like cybersecurity, which monetize AI usage more broadly.
💡 Practical portfolio framing for an under‑$15 AI security name:
Treat it as high‑beta satellite exposure, not a core position
Size it so a major drawdown does not damage overall strategy
Ground the thesis in product‑level ROI—breach reduction, analyst productivity, SOC outcomes—rather than generic “AI” narratives
With AI now shaping GDP, earnings, credit, and geopolitics, Morgan Stanley frames it as a macro force requiring diversified exposure across infrastructure, ROI software, and defensive segments like AI security.[5][12] In that context, a stock like SentinelOne can act as:
A leveraged beneficiary of AI adoption curves
A risk hedge against AI‑driven cyber incidents
A test case for whether AI‑native security can sustain attractive unit economics
⚡ Section takeaway: The opportunity in an under‑$15 AI security stock is meaningful but demands disciplined sizing, multi‑year horizons, and diversification across the AI value chain.
Conclusion: How SentinelOne Fits a Disciplined 2026 AI Strategy
By 2026, Wall Street expects a step‑change in AI capabilities and capital flows, with nearly $3 trillion in infrastructure spend and a pivot from training to inference reshaping where value accrues.[5][11] In that setting, AI‑native cybersecurity platforms like SentinelOne sit where money and risk converge: securing LLM agents, SOC workflows, and high‑value data atop a strained infrastructure layer.[1][2][7]
Research from Morgan Stanley, SANS, and leading AI labs consistently elevates security, governance, and operationalized AI to mission‑critical status.[2][4][5] For investors building an AI portfolio beyond mega‑caps, a sub‑$15 AI security name offers a high‑beta yet fundamentally grounded way to express the AI buildout—tying returns to the enduring need to make each new unit of AI capability both safer and more productive.
Sources & References (10)
1AI Ecosystem Shift: Security and Reliability Take Center Stage This is critical Anand Singh, What we are seeing goes well beyond a simple tooling acquisition. AI companies are no longer just build...
2How to Integrate AI into Modern SOC Workflows Artificial intelligence (AI) is making its way into security operations quickly, but many practitioners are still struggling to turn early experimentation into consistent operational value. This is be...
3Another Wave of AI Disruption Is Coming This Year, Morgan Stanley Says - Business Insider Morgan Stanley says markets are unprepared for AI disruptions in the next few months. Here are its 3 top predictions.
By William Edwards
Mar 11, 2026, 3:52 PM UTC
This story is available exclusivel...4AI Disruption Concerns Do Not Compute AI Disruption Concerns Do Not Compute
Mar 11, 2026
Investors have been punishing software and data stocks on worries about AI disruption, but those fears don’t match the fundamentals.
Author
Dan S...- 5AI Is Now a Macro Variable. Are You Positioned? The Big Picture: AI is no longer a tech story—it is a macro variable influencing GDP, earnings, credit markets and geopolitics at industrial scale. With trillions in infrastructure still to be deploye...
6Prediction: This AI Infrastructure Stock Will Be the Biggest Beneficiary of OpenAI's Growth by End of 2026 | The Motley Fool By Jeremy Bowman — Jan 27, 2026 at 9:45PM EST
Three years after OpenAI launched ChatGPT, the AI boom is in full effect. Stocks like Nvidia have soared, adding trillions of dollars in market value, bu...7Morgan Stanley warns an AI breakthrough Is coming in 2026 — and most of the world isn’t ready Morgan Stanley warns an AI breakthrough Is coming in 2026 — and most of the world isn’t ready
Nick Lichtenberg
Fri, March 13, 2026 at 12:20 AM PDT
A massive AI breakthrough is coming in the first h...8Morgan Stanley warns an AI breakthrough Is coming in 2026 — and most of the world isn’t ready Morgan Stanley warns an AI breakthrough Is coming in 2026 — and most of the world isn’t ready
By Nick Lichtenberg
Business Editor
March 13, 2026, 3:20 AM ET
A massive AI breakthrough is coming in th...9EP 446 : Morgan Stanley Warns: AI Breakthrough in 2026 EP 446 : Morgan Stanley Warns: AI Breakthrough in 2026
AI Brief
Get ready for a transformative leap in AI capabilities, predicted by Morgan Stanley to happen in 2026. With top US AI labs scaling up ...10OpenAI to acquire Promptfoo, a platform focused on testing and securing AI systems OpenAI announced plans to acquire Promptfoo, an AI security platform that helps organizations test and secure AI systems during development.
The tools help developers identify vulnerabilities such as...
Generated by CoreProse in 1m 50s
10 sources verified & cross-referenced 1,544 words 0 false citationsShare this article
X LinkedIn Copy link Generated in 1m 50s### What topic do you want to cover?
Get the same quality with verified sources on any subject.
Go 1m 50s • 10 sources ### What topic do you want to cover?
This article was generated in under 2 minutes.
Generate my article 📡### Trend Radar
Discover the hottest AI topics updated every 4 hours
Explore trends ### Related articles
US Intelligence Elevates AI: Inside the 2026 Global Threat Rankings
Safety#### Inside “Hunter Alpha”: Is DeepSeek Quietly Red‑Teaming the Market?
Safety#### How Litera’s Lito + Midpage Integration Redefines Legal AI Workflows
Safety
About CoreProse: Research-first AI content generation with verified citations. Zero hallucinations.
Top comments (0)