Agentic AI Deals: M&A, Acquisitions, and VC Funding (June 12-19, 2026)

The week of June 12 to 19, 2026 will be remembered as the moment the AI IPO supercycle shifted from speculation to market reality. SpaceX debuted on the Nasdaq under ticker SPCX on June 12, raising a record $75 billion at a $1.77 trillion implied valuation, with shares surging 19% on their first day and touching an all-time high of $225.64 by June 16. One week earlier, Anthropic filed a confidential S-1 with the SEC on June 1, revealing an annualized revenue run rate crossing $47 billion and targeting a $965 billion valuation that leapfrogs OpenAI, which followed with its own confidential S-1 on June 8 at a valuation range of $730 billion to $1 trillion. On the acquisition front, Palo Alto Networks completed its purchase of AI gateway provider Portkey on May 29 in a deal analysts peg at the $700 million class, while Cisco closed its $400 million acquisition of Israeli AI agent security startup Astrix. Cloudflare acquired VoidZero, the company behind the Vite JavaScript toolchain, on June 4 to unify developer tooling with its edge network. Meanwhile, BlackRock, Nvidia, and Microsoft are closing a $40 billion acquisition of Aligned Data Centers as the four major hyperscalers collectively commit $725 billion in 2026 capital expenditures, up 77% year-over-year. CrowdStrike has committed $1.16 billion across two pending acquisitions, SGNL ($740 million for agent identity) and Seraphic Security ($420 million for browser runtime), while Zscaler acquired Symmetry Systems for AI identity governance in May. Fortinet continues its organic AI strategy while its $100 million acquisition of Perception Point for AI-powered SaaS threat protection extends its workspace security coverage. The deal velocity across IPOs, acquisitions, and infrastructure investments points to a single conclusion: capital is consolidating around the companies that control the agentic AI stack, from silicon and data centers through gateways and identity layers to the agent runtime itself.

SpaceX IPO Shatters Records with $75 Billion Raise

SpaceX priced its initial public offering at $135 per share on June 12, selling more than 555 million shares and raising $75 billion in what became the largest IPO in market history. The stock opened above $175, surged past $225 by June 16 to set its all-time high, then settled around $185 by June 19. The $75 billion haul exceeds the combined capital raised by all 71 other IPOs in 2026, and the $1.77 trillion debut valuation clears Tesla's $1.49 trillion market capitalization. SpaceX controls more than 80% of U.S. rocket launches and its Starlink service now reaches over 12 million subscribers across 160 countries, though the company posted a $4.9 billion net loss on $18.7 billion in 2025 revenue.

Why it matters. The SpaceX IPO creates a gravitational reference point for every AI and technology company considering public markets in 2026. Cerebras, which raised $5.55 billion in its May 14 IPO at $185 per share and saw a 68% first-day pop, had already demonstrated that public market appetite for AI infrastructure was robust, but SpaceX's $75 billion raise proves that trillion-dollar-plus tech debuts are viable. This directly de-risks the IPO calculus for both Anthropic and OpenAI, which are now competing not just for model supremacy but for the distinction of being the first pure-play AI company to achieve a trillion-dollar public valuation.

The SpaceX listing also reshapes the competitive landscape for AI infrastructure spending. SpaceX's merger with Elon Musk's xAI means that Starlink's satellite bandwidth and xAI's Grok models now share a corporate treasury with $75 billion in fresh capital. For cybersecurity vendors like Palo Alto Networks, Fortinet, and CrowdStrike that are racing to secure agentic AI workloads, the SpaceX-xAI combination represents both a massive prospective customer and a potential competitor in AI infrastructure services. The broader IPO wave, which includes Cerebras at $5.55 billion and is now followed by the Anthropic and OpenAI filings, signals that public markets are ready to absorb hundreds of billions in AI-related capital over the next twelve months.

Anthropic and OpenAI File Dueling S-1s, Targeting Trillion-Dollar Listings

Anthropic confidentially submitted its draft S-1 registration statement to the SEC on June 1, 2026, targeting an October listing on the Nasdaq at a $965 billion valuation. The company's annualized revenue run rate crossed $47 billion in May, up from roughly $10 billion a year earlier, representing an approximately 4.7x revenue increase in twelve months. Goldman Sachs, JPMorgan, and Morgan Stanley are named as lead underwriters, and the offering is expected to raise more than $60 billion. One week later, OpenAI filed its own confidential S-1 on June 8, disclosing a valuation range between $730 billion and $1 trillion, with Goldman Sachs and Morgan Stanley managing the offering and a dedicated allocation planned for retail investors. OpenAI has not committed to a listing date, noting there are "things we want to do that are likely easier as a private company."

Why it matters. The dueling filings mark the first time two private companies valued at or near one trillion dollars have been simultaneously pursuing IPOs in the same sector. Anthropic's $965 billion valuation leapfrogs OpenAI's last private-round valuation of $852 billion from March 2026, when OpenAI closed a record $122 billion funding round. The valuation inversion signals that investors are now pricing Anthropic's 4.7x revenue growth rate and enterprise positioning as competitively superior, despite OpenAI's larger consumer footprint. For the cybersecurity vendors building agentic security platforms, these filings clarify the foundation model providers that will anchor the next generation of enterprise AI: organizations will need to secure workloads running on both Claude and GPT models, which is precisely the multi-model governance capability that Palo Alto Networks acquired through Portkey and that Check Point is building through its Lakera integration.

The IPO race also has direct implications for the AI gateway market. Both Anthropic and OpenAI will use IPO proceeds to expand compute infrastructure and enterprise sales, which will accelerate enterprise adoption of AI agents. More agents mean more model API calls, more tool invocations, and more token flows, all of which must pass through a governance and security gateway. Cloudflare, IBM DataPower, ServiceNow AI Control Tower, and Kong are all competing for this control plane alongside Palo Alto's Portkey integration into Prisma AIRS. The combined IPO capital from SpaceX, Cerebras, Anthropic, and OpenAI could exceed $200 billion before year-end, creating the largest single-year technology capital injection since the dot-com era and fundamentally reshaping the competitive dynamics for every company in the AI security supply chain.

Palo Alto Networks Completes Portkey Acquisition, Cementing AI Gateway Play

Palo Alto Networks completed its acquisition of Portkey, the San Francisco-based AI gateway provider, on May 29, 2026, after announcing the deal on April 30. While Palo Alto did not disclose the purchase price, analysts at The New Stack characterize it as a "$700M-class AI bet." Portkey processes trillions of tokens per month with the low latency required for agent-to-agent communication, and its AI gateway provides real-time control over AI traffic by centralizing interactions between applications, agents, and AI models. The technology is being integrated into Palo Alto's Prisma AIRS platform. Combined with the $400 million acquisition of Koi Security that closed on April 14, Palo Alto's total 2026 AI-related acquisition spend now exceeds $1.1 billion.

Why it matters. This acquisition positions Palo Alto Networks as the only major security vendor that owns a dedicated AI gateway capable of securing multi-model, multi-agent enterprise deployments. The Portkey integration gives Prisma AIRS visibility into every model request, every agent tool call, and every token flow across an organization, a capability that goes far beyond the endpoint and network security layers where CrowdStrike and Fortinet are primarily competing. CrowdStrike has invested $740 million in SGNL for agent identity and $420 million in Seraphic Security for browser runtime, staking its position at the identity and browser layers. Fortinet is pursuing an organic approach, integrating AI security features directly into FortiAI and FortiSOC with native MCP support, avoiding large acquisitions in favor of leveraging its existing FortiGate G Series data center hardware advantage.

Check Point has taken yet another architectural approach, spending approximately $300 million on Swiss AI security firm Lakera for prompt injection defense, plus an estimated $150 million across three Q1 2026 acquisitions: Cyclops Security ($85 million, attack surface management), Cyata (AI agent identity management), and Rotate (MSP platform). Check Point's total AI security investment exceeds $450 million, but it has no AI gateway asset comparable to Portkey. This means Check Point secures what happens inside the model interaction but lacks the centralized traffic control point that Palo Alto now commands. The strategic question for 2027 is whether Cloudflare AI Gateway, which already processes traffic from 70+ models and recently expanded its partnership with xAI for full Grok integration, will emerge as the neutral, multi-vendor gateway that enterprises prefer over a security-vendor-owned alternative.

Cisco Closes $400 Million Astrix Deal, Google Digests $32 Billion Wiz

Cisco completed its $400 million acquisition of Israeli cybersecurity startup Astrix Security in May 2026. Founded in 2021 by veterans of Israel's elite Unit 8200, Astrix offers software to monitor and secure AI agents, and its capabilities are being integrated into Cisco Identity Intelligence, Duo, and Secure Access. Separately, Google continues to integrate Wiz following the $32 billion all-cash acquisition that closed on March 11, 2026, the largest acquisition in Google's 28-year history. Wiz is maintaining its brand while operating under Google Cloud, providing multi-cloud security capabilities that complement Google's existing Mandiant threat intelligence platform acquired for $5.4 billion in 2022.

Why it matters. Cisco's Astrix acquisition signals that AI agent security has become a standard requirement in enterprise networking platforms, not just a niche for security-first vendors. By integrating Astrix into its zero trust stack alongside the Galileo AI observability platform acquired in April 2026, Cisco is building an agent security layer directly into the network fabric. This positions Cisco to compete with Palo Alto Networks and Zscaler for enterprises that want agent security integrated into their existing network infrastructure rather than bolted on through a separate platform. Zscaler itself acquired Symmetry Systems in May 2026 to add AI identity governance to its Zero Trust Exchange and closed the SquareX browser security deal in February, mirroring CrowdStrike's browser-layer strategy.

Google's digestion of Wiz at $32 billion continues to reshape the cloud security competitive landscape. With Wiz providing cloud security posture management and Mandiant delivering threat intelligence, Google Cloud now offers a security stack that rivals Microsoft's Defender and Sentinel ecosystem. For Fortinet, whose FortiGate-VM integrates with Nvidia BlueField-3 for AI factory security, and Check Point, which is building out its AI Defense Plane, the Google-Wiz combination represents a formidable hyperscaler-native competitor that can bundle security into cloud consumption agreements, applying pricing pressure to standalone security vendors.

The cumulative effect of Cisco at $400 million for Astrix, Google at $32 billion for Wiz, and ServiceNow at $7.75 billion for Armis is that AI agent security is no longer a point solution category. It is being absorbed into the largest enterprise platforms in technology. Standalone security vendors that cannot match this integration depth, whether through acquisitions or organic development, risk being disintermediated by platform vendors that bundle agent security into existing enterprise agreements at marginal cost.

ServiceNow Finalizes $7.75 Billion Armis Acquisition, Plans $12 Billion M&A Spree

ServiceNow completed its acquisition of Armis for approximately $7.75 billion in cash, the largest acquisition in the company's history. The deal, funded through a combination of cash on hand and debt, extends ServiceNow's security platform into the physical and operational layers of the enterprise by adding Armis's cyber exposure management capabilities across OT, IoT, and medical devices. ServiceNow has signaled plans to spend at least $12 billion in total on acquisitions and strategic investments in 2026, having already completed the acquisitions of Veza (AI-native identity intelligence) and Traceloop (code generation) in March.

Why it matters. The Armis acquisition transforms ServiceNow from an IT workflow automation vendor into a full-spectrum security platform that spans IT, OT, IoT, and AI agent infrastructure. ServiceNow's AI Control Tower, which the company offers as a free governance layer for agentic AI, now has Armis's asset discovery feeding it real-time visibility into every connected device and AI agent across the enterprise. This "agent of agents" positioning directly competes with Palo Alto Networks' Prisma AIRS platform, but from the workflow and governance layer rather than the network security layer.

The $12 billion total M&A budget for 2026 dwarfs the acquisition spending of every pure-play security vendor. Palo Alto Networks has spent roughly $1.1 billion on Portkey and Koi combined. CrowdStrike has committed approximately $1.16 billion across SGNL and Seraphic. Check Point has deployed approximately $450 million across Lakera and its three Q1 acquisitions. ServiceNow's spending power, fueled by its $30+ billion subscription revenue base, allows it to acquire capabilities across identity (Veza), asset management (Armis), and observability (Traceloop) simultaneously, building a horizontal security platform that could displace multiple point solutions from Fortinet, IBM, and others. For IBM, which completed its $11 billion Confluent acquisition in December 2025 and announced a $10 billion quantum computing investment in June, ServiceNow's aggressive M&A pace represents a direct challenge in the enterprise workflow and AI governance space where both companies compete.

Cloudflare Acquires VoidZero, Expands xAI Partnership for AI Gateway

Cloudflare acquired VoidZero, the open-source-first company behind the Vite JavaScript build tool, on June 4, 2026. The deal brings the Vite ecosystem, including the Vitest test runner, Rust-based Rolldown bundler, and Oxc toolchain, under Cloudflare's emerging technology organization led by Vite creator Evan You. As part of the acquisition, Cloudflare committed $1 million to an independent Vite ecosystem fund. On the same day, Cloudflare expanded its partnership with xAI for full Grok model integration into Cloudflare AI Gateway, adding to its existing support for 70+ models.

Why it matters. While VoidZero is a developer tooling acquisition rather than a security deal, it strengthens Cloudflare's position as the default deployment platform for AI-native web applications, which in turn feeds traffic through its AI Gateway. By controlling both the build toolchain (Vite) and the runtime edge (Workers), Cloudflare creates a frictionless path from local development to global deployment, with the AI Gateway sitting at the center of every model API call. The xAI Grok integration further positions Cloudflare's AI Gateway as the most model-agnostic option in the market, a contrast to Palo Alto Networks' Portkey, which is being folded into the security-specific Prisma AIRS platform.

Cloudflare's developer-up strategy differs fundamentally from the enterprise-down approaches of Palo Alto Networks, Check Point, and ServiceNow. Rather than selling AI governance to CISOs, Cloudflare embeds gateway controls into the developer workflow. This positions Cloudflare to capture the long tail of AI applications that enterprises build internally, applications that may never touch the enterprise security platform but still need rate limiting, model routing, and observability. For Fortinet, which has deep relationships with network operations teams but limited developer platform penetration, Cloudflare's AI Gateway expansion represents a competitive threat at the application layer that FortiAI's MCP support alone may not address.

CrowdStrike and Zscaler Double Down on Identity and Browser Layers

CrowdStrike has committed $1.16 billion across two major pending acquisitions that are expected to close in the company's fiscal Q1 2027. The $740 million deal for SGNL, the identity security startup that provides fine-grained authorization for AI agents, and the $420 million acquisition of Seraphic Security for browser runtime protection together represent CrowdStrike's bet that the identity and browser layers are the critical control surfaces for agentic AI. In parallel, Zscaler acquired Symmetry Systems in May 2026 to bolster AI identity governance within its Zero Trust Exchange, adding to the SquareX browser security acquisition that closed in February. Zscaler also expanded a strategic partnership with Saviynt in June to enhance identity-centric zero trust, while Saviynt itself closed a $700 million Series B earlier this year.

Why it matters. CrowdStrike and Zscaler are converging on the same architectural thesis: that securing AI agents requires controlling both who the agent claims to be (identity) and where the agent executes (browser and runtime). CrowdStrike's combined $1.16 billion in identity and browser acquisitions positions its Falcon platform as the agent-aware successor to traditional endpoint detection, while Zscaler's Symmetry Systems integration brings identity-data mapping into its cloud-native zero trust architecture. The identity layer is particularly critical because AI agents authenticate to enterprise systems using service identities, API keys, and OAuth tokens that traditional identity governance platforms were never designed to manage at agent scale.

This identity-centric approach contrasts sharply with Palo Alto Networks' gateway-centric strategy and Fortinet's infrastructure-first model. Where Palo Alto controls the traffic and Fortinet controls the data center perimeter, CrowdStrike and Zscaler are trying to control the agent itself. Check Point's acquisition of Cyata, which specializes in managing identities and permissions of AI agents operating in the enterprise, shows that even the historically perimeter-focused Israeli vendor recognizes the agent identity gap.

The vendor that solves agent identity at scale, providing continuous authorization rather than point-in-time authentication, will likely command the largest share of the estimated $3 trillion in agentic AI productivity gains that analysts project annually. IBM, with its watsonx Orchestrate platform and DataPower Nano Gateway, is approaching the same problem from the enterprise middleware layer, leveraging its $11 billion Confluent acquisition for real-time event streaming that feeds agent authorization decisions.

Hyperscaler Capex Hits $725 Billion as AI Infrastructure Race Accelerates

The four major hyperscalers have collectively committed $725 billion in 2026 capital expenditures, up 77% from the record $410 billion spent in 2025. Amazon leads with approximately $200 billion, followed by Alphabet at $175 billion to $185 billion, Meta at $115 billion to $135 billion, and Microsoft at $110 billion to $120 billion. The spending funds Nvidia GPUs, custom silicon, data center construction, and power infrastructure. Separately, the BlackRock-Nvidia-Microsoft-xAI consortium is closing a $40 billion acquisition of Aligned Data Centers, the first transaction for the AI Infrastructure Partnership formed last year. Aligned operates 50 data center sites with 5 GW of available capacity across the U.S. and Latin America, and the consortium plans to provide $30 billion in equity with the possibility of expanding to $100 billion including debt.

Why it matters. The $725 billion in hyperscaler capex creates the physical infrastructure layer upon which every agentic AI workload will run, and the security vendors that can embed themselves into this infrastructure at deployment time will capture durable, recurring revenue. Fortinet is already positioned here through its FortiGate-VM integration with Nvidia BlueField-3 DPUs for AI factory security, giving it a hardware-adjacent wedge into hyperscaler data centers that neither Palo Alto Networks nor CrowdStrike currently matches. IBM's $10 billion quantum computing investment, announced June 2, represents a parallel bet that next-generation compute architectures will require purpose-built security and data processing layers, areas where its DataPower gateway and watsonx platforms already have footholds.

The Aligned Data Centers deal at $40 billion is particularly significant because it brings Nvidia, Microsoft, and xAI together as co-investors in physical AI infrastructure. When the same companies that build the chips, run the cloud, and train the models also own the data centers, the entire AI stack from silicon to software becomes vertically integrated. For enterprise customers, this raises questions about vendor lock-in and data sovereignty that security platforms must address. Cloudflare, with its distributed edge architecture across 330+ cities, offers a counterpoint to hyperscaler centralization, and its AI Gateway serves as a model-routing layer that can spread workloads across providers rather than concentrating them in a single hyperscaler's infrastructure.

The sheer scale of infrastructure investment also creates a massive addressable market for security vendors. AWS announced Continuum, a new security offering for code vulnerabilities in June, and secured a $2.6 billion DHS cloud contract, demonstrating that hyperscaler security capabilities are expanding in parallel with their infrastructure. Microsoft is debuting a multi-model AI security scanner in preview this month through its Agent 365 platform. Every dollar of the $725 billion in capex represents infrastructure that must be secured, monitored, and governed, which is why CrowdStrike partnered with CoreWeave and Nvidia for Charlotte AI AgentWorks, and why Fortinet has invested in FortiGate integration with Nvidia DPUs at the data center hardware layer.

Numbers at a Glance

SpaceX raised $75 billion in the largest IPO ever at a $1.77 trillion valuation, with shares touching $225.64 before settling around $185 by June 19. Anthropic filed its S-1 at a $965 billion valuation with $47 billion in annualized revenue, while OpenAI filed at a $730 billion to $1 trillion range one week later. Palo Alto Networks has spent over $1.1 billion on AI acquisitions in 2026 across Portkey and Koi Security ($400 million). CrowdStrike has committed $1.16 billion across SGNL ($740 million) and Seraphic Security ($420 million). ServiceNow completed its $7.75 billion Armis acquisition and plans $12 billion in total 2026 M&A spending. Cisco closed the $400 million Astrix deal for AI agent security. Check Point has invested approximately $450 million across Lakera ($300 million) and three Q1 acquisitions totaling $150 million. Fortinet has acquired Perception Point for $100 million to extend AI-powered threat protection across SaaS collaboration tools. Google continues integrating Wiz following the $32 billion close in March. Cerebras raised $5.55 billion in its May 14 IPO at $185 per share. Hyperscaler capex stands at $725 billion combined for 2026, led by Amazon at $200 billion and Alphabet at $175 billion to $185 billion. The BlackRock-Nvidia-Microsoft consortium is closing the $40 billion Aligned Data Centers acquisition. Four of the five largest venture rounds in history closed in Q1 2026 alone: OpenAI ($122 billion), Anthropic ($30 billion), xAI ($20 billion), and Waymo ($16 billion). Total cybersecurity M&A value across all vendors tracked in this report now exceeds $55 billion year-to-date, driven by the race to own every layer of the agentic AI security stack.

References

1. CNBC: SpaceX IPO takeaways: SPCX closes at $161, jumping 19% after record debut
2. Yahoo Finance: SpaceX Raises Record $75 Billion in Historic IPO
3. CNBC: Anthropic tops OpenAI as most valuable AI startup, nears $1 trillion valuation
4. Fortune: OpenAI files confidential SEC S-1 paperwork for IPO
5. The New Stack: Palo Alto Networks makes a $700M-class AI bet on Portkey gateway
6. Palo Alto Networks: Completes Acquisition of Portkey to Secure AI Agents
7. Calcalist: Palo Alto Networks completes $400 million acquisition of Koi
8. Calcalist: Cisco acquires AI security startup Astrix for $400 million
9. ServiceNow Newsroom: ServiceNow completes Armis acquisition
10. TechCrunch: Google wraps up $32B acquisition of cloud cybersecurity startup Wiz
11. Cloudflare: Acquires VoidZero to Build the Future of the AI-Native Web
12. Tom's Hardware: Google, Microsoft, Meta, and Amazon capex spending to hit $725 billion in 2026
13. Data Center Knowledge: Nvidia-Backed Group Drops $40B on Aligned Data Centers
14. CNBC: Cerebras (CBRS) starts trading on Nasdaq after IPO
15. SecurityWeek: CrowdStrike to Acquire Browser Security Firm Seraphic for $420 Million
16. Crunchbase News: Venture Funding To Foundational AI Startups In Q1 Was Double All Of 2025