Palantir (PLTR) on Solana

Palantir CEO Alex Karp has argued that enterprises are "paying for tokens that create no value" and unwittingly exposing their competitive edge to AI providers who end up "stealing the weights and alpha" of their business. Microsoft CEO Satya Nadella has now put forward a parallel formulation he calls the "reverse information paradox": organizations "pay for intelligence twice — once with money, and again with something even more valuable: the proprietary knowledge you must reveal." Writing in a recent blog post, Nadella observed that frontier models absorb internal prompts, agent tools, and user corrections, effectively converting enterprise institutional knowledge into model improvements — "every correction is distilled into institutional know-how."

The alignment between two of the most prominent voices in enterprise tech reinforces Palantir's core pitch. The company's Ontology layer is designed to connect AI models to business operations while controlling data exposure — a direct architectural answer to the risk both CEOs describe. Karp has framed Palantir's approach as making AI "safe and useful and precise" without allowing providers to exploit the proprietary data customers feed into their systems.

Apple filed suit on July 10 alleging that OpenAI stole its trade secrets through former employees Chang Liu and Tang Yew Tan to develop consumer hardware, with the complaint stating directly that "OpenAI has been stealing Apple's trade secrets and confidential information." The lawsuit lends concrete weight to warnings Palantir CEO Alex Karp had previously made in a CNBC Squawk Box interview, where he identified intellectual property theft as a structural risk of working with frontier large language model providers, asking "Why are they charging for tokens if it's so valuable?" and accusing AI companies of creating what he called "a wealth tax" on customers.

Karp has positioned Palantir's Ontology product as the enterprise answer to those risks, arguing it "makes it safe and useful and precise" while preventing LLMs from caching customer data or transferring their IP. Apple's legal action against OpenAI — one of the most prominent AI providers in the world — substantiates the concern that enterprise and consumer customers face genuine IP exposure when engaging with frontier AI systems, a dynamic that Palantir's security-first approach to AI deployment is explicitly designed to address.

Palantir CEO Alex Karp told CNBC's Squawk Box on July 1 that "something has gone completely wrong" with OpenAI and Anthropic, centering his critique on their token-based pricing models. Karp argued that enterprise customers have grown fatigued by per-token costs, characterizing the prevailing sentiment as an unwillingness to accept the model's economics, and said businesses increasingly want control over "their compute, their models, their data stack and their alpha" — effectively ownership of the AI production stack rather than dependence on closed API providers.

The remarks sharpen Palantir's competitive positioning at a moment when the company is leaning hard into data sovereignty as a differentiator. Karp pointed to Palantir's expanded Nvidia partnership for custom government AI models as evidence that customers are moving toward open-weight deployments with deterministic cost structures. The argument has commercial backing: Palantir reported U.S. commercial revenue of $595 million in Q1 FY2026, a 133% year-over-year increase, while overall revenue grew roughly 85%, suggesting enterprises are already voting with contracts for the sovereignty-first model Karp is promoting.

Palantir Technologies and SNP SE have announced a strategic partnership to develop AI-powered solutions for enterprise SAP transformations. The collaboration pairs SNP's specialized SAP data migration software — including its Kyano® platform — with Palantir's AI and data infrastructure, targeting large-scale migrations to SAP Cloud ERP environments. SNP brings an installed base of more than 3,000 customers and a record of 15,000 completed SAP transformation projects to the deal.

The first joint offering, called Test Data Proposal, uses AI to automatically identify relevant test data sets for SAP migration test cases, replacing a process that has traditionally been done manually. Both companies framed the partnership around improving speed, compliance, and predictability for enterprises modernizing their SAP landscapes, with Palantir's Head of US Commercial saying it would help customers transform their SAP environments in a controlled and compliant way.

Palantir is facing renewed political scrutiny in the United Kingdom over its role managing the NHS Federated Data Platform, the critical data infrastructure underpinning the National Health Service. Critics are questioning the UK's reliance on a U.S. technology provider for such sensitive public health data, raising concerns about contract continuity at the next renewal point. A former NHS data leader noted that replacing Palantir before the next contractual break point is unlikely, suggesting the renewal window—rather than an immediate termination—represents the primary risk to the relationship.

The pressure adds a political dimension to Palantir's UK government revenue that goes beyond standard commercial risk. While the contract appears stable in the near term, sustained scrutiny could complicate future renewals and affect how UK government bodies approach other engagements with Palantir. The situation also underscores the broader exposure government-focused technology firms face when foreign-owned platforms become embedded in sovereign public infrastructure.

Palantir and Nvidia have announced a formal partnership to bring sovereign AI capabilities to U.S. federal agencies and critical infrastructure operators. The alliance combines Nvidia's open-source Nemotron language models with Palantir's core software stack—AIP, Foundry, Ontology, and Apollo—enabling government customers to train and operate AI systems entirely within controlled environments, with no external data sharing required. The integration supports deployment in classified and air-gapped settings, with explicit data authorization controls, secure environment isolation, and full data erasure capabilities. CEO Alex Karp described the combination as allowing the U.S. government to "unleash the full power of LLMs while removing underlying security risks," while Nvidia CEO Jensen Huang positioned open-source AI as foundational to national security and U.S. technological leadership.

The announcement follows Palantir's strongest reported quarter as a public company, with Q1 2026 revenue up 85% year-over-year, U.S. business revenue rising 104%, and government revenue growing 84% annually. Military usage of the company's Maven Smart System quadrupled over the same period. The partnership rollout includes deployment engineering support and production fine-tuning tools through Nvidia NIM microservices, deepening Palantir's position as a primary AI infrastructure layer for the national security sector.

Palantir (PLTR) has shed roughly 24% year-to-date in 2026, tumbling from a 52-week high of $207.52 to a low of $122.68, and the stock's retreat has intensified the debate between bulls and bears over whether the selloff reflects a genuine inflection in the business or simply the unwinding of last year's crowded trade. At 86.91x forward earnings and 39.74x sales, the valuation remains rich relative to software peers even after the pullback, and bears argue the multiple still prices in years of near-perfect execution — leaving minimal margin for any stumble on international expansion, government contract renewals such as the U.K. NHS deal, or competitive pressure from OpenAI, Anthropic, and Databricks moving into Palantir's territory.

Bulls counter that the underlying fundamentals have not cracked: Q1 2026 revenue hit a record $1.63 billion, up 85% year-over-year, marking eleven consecutive quarters of accelerating growth, with operating cash flow nearly tripling to $899 million. Full-year 2026 guidance calls for $7.656 billion in revenue (+71%), and the customer base has topped 1,000 accounts with net dollar retention at 150%. Among 28 analysts tracked, 19 rate PLTR a strong buy with an average price target of $194.81, implying roughly 45% upside from recent levels. The central disagreement is whether the AI Platform (AIP) can sustain that growth cadence long enough to grow into its valuation, or whether expectations set during last year's rally have simply become too elevated for anything short of exceptional execution to satisfy.

Palantir Technologies was selected on June 23, 2026 to serve as the cloud data layer for the US Army's Next Generation Command and Control (NGC2) program, with the company's Palantir Foundry platform providing the foundational data architecture. NGC2 is the Army's initiative to build a modern, data-driven command and control system by integrating commercial and government technology solutions, aiming to compress the timeline between experimentation and fielding mission-ready capabilities.

Palantir's role in NGC2 pairs its cloud infrastructure with Anduril's Lattice tactical data layer, creating an extensible system designed for rapid scaling and interoperability across diverse software and applications. The win deepens Palantir's foothold in large-scale US defense programs and underscores the company's strategy of embedding its data platforms at the infrastructure level of critical military systems, though specific contract value was not disclosed.

Palantir (PLTR) fell nearly 6% to a 52-week low of $106.38 on Thursday, extending a seven-session losing streak and putting the stock on track for its worst week in over five years. The decline is down to two converging pressures: mounting skepticism over the company's rich valuation and a fresh legal setback in Europe. With a forward price-to-earnings multiple of around 120x — well above most enterprise software peers — investors appear to be rotating toward AI infrastructure plays such as chip and server suppliers, which are seen as having cleaner near-term revenue visibility compared to software companies still navigating longer enterprise-adoption cycles. PLTR is now down roughly 39% year-to-date.

Adding to the pressure, Palantir lost a legal case in a Zurich court involving Swiss investigative outlet Republik, a ruling that raised new questions about its ability to win government contracts in Europe and clouded its international growth prospects. Together, the valuation re-rating and the European legal headwind have accelerated selling despite retail sentiment on the stock remaining broadly bullish.

Michael Burry's Scion Asset Management covered approximately half of its Palantir short position at $107.15 per share, according to recent disclosures. The move represents a partial profit-taking on a trade that has been well-timed: PLTR is down roughly 42% year-to-date, making the short bet profitable before the trim.

Burry did not fully exit his bearish thesis — he continues to hold put options on Palantir, signaling he expects further downside even after locking in partial gains. The tactical adjustment suggests a measured reduction in risk rather than a conviction change, with retail sentiment on platforms like Stocktwits running bullish against his still-cautious positioning.