AI models will quickly automate the majority of expert work, but they will struggle with the final, most complex 25%. For a long time, human expertise will be essential for this 'last mile,' making it the ultimate bottleneck and source of economic value.

AI models lack access to the rich, contextual signals from physical, real-world interactions. Humans will remain essential because their job is to participate in this world, gather unique context from experiences like customer conversations, and feed it into AI systems, which cannot glean it on their own.

AI tools can now perform complex fundamental analysis, commoditizing a once-essential analyst skillset. This shift means that a deep understanding of market structure, positioning, and trading dynamics is becoming the more valuable and differentiating skill for portfolio managers seeking an edge.

As AI makes complex financial data and analysis a commodity for both bankers and their clients, the key differentiator will no longer be information. Bankers will have to provide value through human-centric skills: understanding psychology, navigating boardroom tactics, and providing judgment that a machine cannot replicate.

The transition from human to machine-driven trading has a specific threshold: one-tenth of a second, the lower limit of human time perception. Once trading speeds crossed this barrier, human decision-making became too slow to compete, necessitating algorithmic control for execution.

As AI masters the analysis of financial filings and transcripts, the source of investment alpha may shift to information that is difficult for models to process. Qualitative insights from attending conferences, judging a CEO's character via a handshake, or other forms of scuttlebutt could become increasingly valuable differentiators for human investors.

Unlike highly electronic corporate bond markets, the leveraged loan market remains manually traded and is a key point of fragility. With 15% of the universe exposed to tech, the uncertainty around AI disruption could cause liquidity to evaporate quickly for companies deemed 'high risk'.

The future of AI in finance is not just about suggesting trades, but creating interacting systems of specialized agents. For instance, multiple AI "analyst" agents could research a stock, while separate "risk-taking" agents would interact with them to formulate and execute a cohesive trading strategy.

To compete with superior AI, a human trader should focus on building a portfolio of undervalued assets today. When hyper-intelligent AIs eventually arrive and re-price markets efficiently, they will likely buy the very assets the human trader already holds, validating the initial thesis and accelerating gains.