Review date: June 14, 2026
The AI Portfolio remains a deliberately focused thematic portfolio built around the artificial-intelligence value chain. The current holdings give the portfolio exposure to three different layers of the AI opportunity set: Innodata for AI data and model-evaluation services, Penguin Solutions for AI infrastructure and memory-centric systems, and Camtek for semiconductor inspection and metrology tied to advanced packaging, HBM, and AI-chip production.
Current positioning
The portfolio currently holds three open positions: Innodata Inc. (INOD), Penguin Solutions, Inc. (PENG), and Camtek Ltd. (CAMT). Based on stored quote-history performance, INOD is down 6.81% from its first quote, while PENG and CAMT are newly added and flat versus their stored initial quotes. As of the latest market data available for June 12, 2026, INOD was quoted at $100.90, PENG at $64.13, and CAMT at $193.27.
This is a concentrated AI portfolio, but the concentration is spread across different business models rather than three companies doing the same thing. INOD is tied to the data, evaluation, and agentic-AI enablement layer. PENG is tied to AI factories, advanced compute, and memory infrastructure. CAMT is tied to the physical semiconductor supply chain that supports AI accelerators, HBM, chiplets, and advanced packaging.
Recent performance drivers
The main short-term drag is INOD, which has pulled back despite strong recent operating momentum. Innodata reported first-quarter 2026 revenue of $90.1 million, up 54% year over year, adjusted EBITDA of $25.0 million, and raised full-year 2026 revenue-growth guidance to approximately 40% or more. The company also cited new Big Tech engagements expected to generate about $51 million of 2026 revenue, which keeps the AI data-services thesis intact. ([investor.innodata.com](https://investor.innodata.com/news/news-details/2026/Innodata-Reports-Record-First-Quarter-2026-Results/default.aspx?utm_source=openai))
PENG was added as an AI infrastructure holding. Penguin Solutions reported fiscal second-quarter 2026 net sales of $343 million, and management raised its full-year net sales and EPS outlook in April. On June 1, 2026, the company also reaffirmed its fiscal 2026 outlook and said it expected both net sales and diluted EPS to be at the high end of prior guidance ranges, supported by strong agentic-AI-driven demand across Integrated Memory and AI Infrastructure. ([ir.penguinsolutions.com](https://ir.penguinsolutions.com/news/news-details/2026/Penguin-Solutions-Reports-Q2-Fiscal-2026-Financial-Results/default.aspx?utm_source=openai))
CAMT was added as a semiconductor-equipment and advanced-packaging exposure. Camtek reported first-quarter 2026 revenue of $121.7 million and said it expected second-half 2026 revenue growth of more than 25% versus the first half, based on strong order momentum. The company also announced more than $105 million of new multi-system orders on June 2, 2026, including a $55 million order from a tier-1 OSAT supporting AI applications and more than $50 million of Hawk-system orders from a leading HBM player for AI-related applications. ([camtek.com](https://www.camtek.com/news-and-events/camtek-announces-results-for-the-first-quarter-of-2026/?utm_source=openai))
Risk concentration
The portfolio is small and concentrated, so single-stock volatility will be high. The shared macro risk is that AI capital spending slows, customers delay infrastructure deployments, or valuation multiples compress across AI-linked equities. Because all three holdings are tied to AI adoption, the portfolio may perform poorly if investor sentiment rotates away from high-growth AI beneficiaries.
Company-specific risks also differ. INOD carries customer-concentration and project-timing risk, especially because AI data and evaluation contracts can scale quickly but can also be reprioritized by large technology customers. PENG carries execution risk around AI factory deployments, memory demand, customer concentration, and the competitive market for AI infrastructure solutions. CAMT carries semiconductor-cycle risk, order-timing risk, and exposure to advanced-packaging capacity cycles, even though current HBM and AI-related demand remain supportive.
Close or hold?
No positions should be closed this week. All three open positions are protected by the minimum holding-period rule and are far younger than 180 days. More importantly, there is no clear thesis break, catalyst failure, deteriorating fundamental signal, liquidity issue, or portfolio-fit problem that would justify an override. INOD is down, but a temporary drawdown is not enough reason to close a position when the underlying AI-growth thesis remains active.
- INOD: Hold. The pullback should be monitored, but recent revenue growth, margin expansion, raised guidance, and Big Tech engagement support the thesis.
- PENG: Hold. The position is new, and recent management commentary supports the AI infrastructure and memory-demand thesis.
- CAMT: Hold. The position is new, and recent AI-related order activity supports the advanced-packaging inspection thesis.
What to watch next
For INOD, the next key test is whether the company converts its 2026 guidance into durable, diversified revenue rather than one-time project spikes. Watch Big Tech customer concentration, gross margin stability, and whether new AI evaluation and observability offerings gain broader commercial traction.
For PENG, watch the next earnings update for evidence that AI/HPC pipeline strength is becoming recognized revenue, not just backlog or sales commentary. The portfolio also needs confirmation that margin performance can improve as the company scales AI infrastructure and integrated-memory offerings.
For CAMT, the key watch items are order conversion, delivery timing, and whether HBM and advanced-packaging demand remain strong into 2027. The June 2026 order announcements are encouraging, but the stock’s valuation and recent momentum mean expectations are likely elevated.
Bottom line
The AI Portfolio remains well aligned with its artificial-intelligence mandate. It is not a broad market portfolio; it is a focused basket of companies positioned around AI data services, AI infrastructure, memory systems, HBM, and advanced semiconductor packaging. That focus creates upside potential if AI investment continues, but it also increases volatility. This week, the disciplined decision is to hold all three positions and allow the theses more time to develop.
Risk disclaimer: This article is for informational and editorial portfolio-review purposes only and is not financial advice, investment advice, or a recommendation to buy or sell any security. The AI Portfolio may hold volatile equities, and thematic AI stocks can experience sharp losses. Investors should conduct their own research and consider their risk tolerance before making investment decisions.