AI Content Is Not the Problem

The argument against AI-generated content mistakes the tool for the thinking. A fund manager who arrives at a genuine position, a view on market structure, a critique of how the industry communicates, an observation about what clients actually need, and uses AI to shape and execute that position is not producing artificial content. The intelligence is human. The typing is not. That distinction matters.

Good content has always required two things: something worth saying, and the craft to say it well. Writers, editors, and copywriters have always supplied the second thing on behalf of people who had the first. AI does the same job. The output is only as good as the thinking that precedes it, which means the responsibility for having a point of view sits exactly where it always did.

For fund managers specifically, the constraint has rarely been execution. Most firms can produce polished prose. The constraint is specificity, a willingness to say something particular about markets, clients, or the industry that could not have been written by a competitor. AI does not change that constraint. It just removes the production cost that was previously used to avoid confronting it.

The firms using AI well are the ones who came to it with something to say. The tool is irrelevant to that condition.

Q: Does using AI to produce content undermine authenticity?

A: Only if authenticity is defined as unassisted writing, which it never has been in a professional context. Executives have used ghostwriters, communications teams, and editors for as long as organisations have published content. The relevant question is whether the ideas, positions, and judgments in the content are genuinely held by the person or firm behind them. If they are, the production method is immaterial.

Q: What does good AI-assisted content actually require from the human?

A: A specific argument, genuine industry knowledge, and a willingness to take a position that could be disagreed with. AI can shape, structure, and execute. It cannot supply the underlying view. Content that lacks a specific human perspective produces generic output regardless of how sophisticated the tool is.

Q: Is AI-assisted content detectable, and does that matter?

A: Detection tools exist but are unreliable, and the question is largely beside the point in a professional publishing context. What sophisticated readers, including financial advisers and asset consultants, actually notice is whether content contains specific, credible, original thinking. Content that does will be read and valued. Content that does not will be ignored, whether it was written by a person or a machine.

Q: How should fund managers think about disclosing AI use in content production?

A: There is no current regulatory obligation in Australia to disclose AI assistance in marketing content, and disclosure norms are still forming. The more relevant consideration is whether the content accurately represents the firm's views and capabilities. Content that does is appropriate regardless of how it was produced. Content that overstates expertise or makes claims the firm cannot support is a compliance issue independent of the production method.

Q: Does AI-assisted content perform differently in search or generative engine results?

A: Not inherently. Search and AI retrieval systems evaluate content against authority and relevance signals, not production method. Well-argued, specific content from a credible domain performs well. Generic content from the same domain performs poorly. The production tool does not change that relationship.