How AI Is Changing Capital Allocation — And Where David Bellairian Draws the Line

By David Bellairian, Founder of Glencore Associates

AI has moved from a novelty in investment management to something close to standard practice, particularly among family offices and privately held capital structures — the category Glencore Associates operates in. It's worth being specific about what that actually means in practice, because the honest answer is narrower and less dramatic than the popular framing of "AI trading algorithms" usually suggests.

What AI is actually doing in this space

According to PwC's 2026 analysis of family office operations, the clearest gains so far are in speed, not decision-making itself: AI is being used to process documents, contracts, and company data "within minutes rather than days, identifying critical terms, highlighting inconsistencies, and flagging potential risks" — work that used to take a due-diligence team days now takes minutes. The same analysis points to cash-flow forecasting and scenario modeling across multi-entity portfolios as an area where AI has meaningfully compressed the time between a question and an answer, giving decision-makers faster visibility across everything they hold.

Morgan Lewis's 2026 guidance on AI adoption in family offices draws a clear line around where that's appropriate: research gathering, summarization, preliminary diligence review, and administrative work are described as lower-risk, well-suited applications, while investment decisions themselves, tax planning, and anything involving material nonpublic information are flagged as areas warranting real caution and continued human control.

What this looks like at Glencore Associates

In that same spirit, David Bellairian incorporates AI-assisted analysis into how he evaluates capital allocation at Glencore Associates — using it to process information faster and surface things worth a closer look, not to make decisions on its own. That might mean working through data on a potential real estate acquisition, screening public information relevant to an equity position, or organizing research across the firm's different asset categories more quickly than would be possible manually.

What AI doesn't do at Glencore Associates is make the actual allocation decision. Every point of research above is exactly that — research — and the judgment about what to do with it stays with the person holding the capital. That's consistent with how the space as a whole talks about this: PwC's framing is that "AI's most important impact is human, not technical," and that human oversight of anything touching an actual investment, tax, or legal decision remains essential, not optional.

Why this matters more for privately held capital

For a fund raising money from outside investors, an AI-driven "edge" is often part of the pitch — something to point to when explaining why a strategy should attract capital. Glencore Associates isn't in that position: it doesn't solicit outside investment or manage money on behalf of others, so there's no pitch to make and no performance claim to substantiate for anyone but the person whose capital is actually at stake. That removes a particular kind of pressure — to overstate what a tool does, or to imply a system is more autonomous or more proven than it actually is — that shows up often in how AI and trading get talked about publicly.

The honest summary

AI, in this context, is a research and efficiency tool — genuinely useful for moving faster through information, and genuinely not a substitute for the judgment involved in deciding what to actually hold. Anyone evaluating a claim about "AI trading" or "algorithmic strategy," from Glencore Associates or anywhere else, is well served by asking the same question Morgan Lewis's guidance implicitly raises: is this tool helping someone decide faster, or is it being presented as the decision-maker itself? Those are very different claims, and only one of them tends to hold up.

This article is for informational purposes only and does not constitute investment advice. It describes a general approach to using AI-assisted analysis and makes no claim about performance or results. Glencore Associates does not manage capital on behalf of outside investors, does not offer any tool or system to the public, and is not currently open to outside investment.

---

About David Bellairian

David Bellairian is the founder of Glencore Associates, where he personally allocates capital across public equities, gold reserves, income-producing real estate, and a commercial vehicle portfolio. He is also the founder and CEO of AIDiscover, an AI-visibility and brand-discovery agency in Los Angeles. Connect with David on LinkedIn, X, or Instagram.

---

Sources referenced in this article:

• PwC, "How AI is Reshaping the Modern Family Office" (2026)

• Morgan Lewis, "AI and the Family Office: Adoption, Investment, and Risk Management" (2026)