AI won’t replace traders—it will redefine their value
Few topics have dominated industry conversations over the last two years quite like artificial intelligence. Yet after moderating a recent panel discussion with market participants from across the buy side, sell side and technology ecosystem, I came away with a slightly different perspective than the one often reflected in headlines.
For all the discussion about AI transforming financial markets, the most interesting question raised by the panel was not whether AI will replace traders, but where human judgement adds the greatest value. Ironically, the more capable AI becomes, the more important that question may become.
Across the discussion, participants agreed that AI is already creating meaningful value. But not in the way many predicted. Rather than replacing decision-makers, AI is increasingly helping firms access information faster, reduce operational friction and focus expertise where it has the greatest impact.
Beyond efficiency: Where AI is creating value today
One theme emerged consistently throughout the discussion: AI's biggest impact today is not full automation. It is workflow enhancement.
The panel highlighted examples ranging from helping traders classify high-touch and low-touch workflows, to connecting datasets, surfacing relevant information and streamlining decision-making. These are not necessarily the headline-grabbing use cases often associated with AI, but they are delivering tangible value today.
What struck me most was how quickly the conversation moved beyond efficiency. Several panellists argued that as AI reduces the cost of accessing information, competitive advantage increasingly shifts from finding data to interpreting it.
That observation has important implications for fixed income markets. If participants have access to similar technologies and increasingly similar information, differentiation comes less from data retrieval and more from context, experience and risk appetite.
In other words, AI is creating a judgement premium.
The discussion also highlighted how firms are adapting talent strategies in response. We discussed increased investment in trader upskilling, AI training programmes and the integration of technology expertise directly within trading teams. Rather than reducing the importance of people, AI may actually increase the demand for individuals capable of combining market expertise with technical fluency.
This was a point highlighted by Alexandre Guignot, European Head of E-Trading at BNP Paribas, who observed:
"What I haven't seen materialise yet is how we transform the trading desk cockpit and how we make use of those tools, to the full power."
Guignot's comment resonated because it captures where the industry stands today. AI is already delivering meaningful benefits in areas such as price discovery, market making and workflow automation. What remains less clear is how these capabilities ultimately reshape the trader's role and day-to-day workflow.
The panel's answer was largely consistent: the value of the trader increasingly lies not in gathering information, but in understanding what it means.
This was particularly evident in discussions around credit markets, where fragmented liquidity, infrequently traded securities and evolving market conditions create nuances that cannot easily be reduced to a rules-based process. AI can narrow the choices. It can identify patterns. It can surface opportunities and risks. But the final decision still requires judgement.
Generative AI and the future trading workflow
Another recurring theme was how dramatically the arrival of generative AI has changed expectations around technology.
For years, industry conversations focused heavily on data infrastructure, analytics and connectivity. Today, the focus is increasingly on accessibility.
As Andy Mahoney, Managing Director, EMEA at FlexTrade Systems, observed:
"AI has reopened doors for clients to be able to build things on the fly that previously would have taken client service interaction or a coder to do so."
That comment captured one of the most important shifts discussed by the panel. The issue is no longer simply whether firms have access to data. The challenge is how efficiently users can reach it, interrogate it and act on it.
AI is lowering the barrier between people and technology. Tasks that once required specialist coding skills, database knowledge or support teams can increasingly be completed by end users themselves.
Interestingly, this does not mean traditional trading tools are disappearing. The discussion highlighted that traders still value familiar interfaces for monitoring markets and executing trades. Where AI is delivering the greatest value is in bringing together information from multiple sources and helping users perform complex tasks that previously required significant manual effort.
The evolution resembles a shift from integrated platforms to connected ecosystems. Just as consumers move seamlessly between specialised applications on their phones, trading workflows could increasingly be assembled from interoperable services, data sources and execution tools linked through APIs. This is consistent with a broader industry push towards creating a more data-rich, user-friendly environment across the trade lifecycle.
Several panellists suggested that APIs are increasingly moving from connectivity tools to strategic infrastructure, enabling firms to build their own workflows, decision layers and user experiences around a common set of services and data.
Trust: The foundation of AI adoption
A recurring tension throughout the discussion centred on trust.
Beneath the conversations about AI, automation and APIs sat a more fundamental question: how confident are we in the decisions being made?
As Max Callaghan, Head of Sales EMEA at MarketAxess, put it:
"Trust and communication is very commercially challenging."
That insight resonated because it extends far beyond AI.
AI requires trust. Electronic markets require trust. Data requires trust. Execution quality requires trust.
The more decision-making becomes data-driven, the more important data quality becomes. AI can accelerate analysis, but it cannot compensate for incomplete, inconsistent or misleading inputs. Several panellists highlighted challenges around dealer selection, market signalling and execution quality, questioning whether increasing transparency and richer performance data may ultimately prove more valuable than traditional liquidity signals alone.
What became clear during the discussion is that firms are unlikely to automate critical decisions simply because they can. They will automate them when they trust the underlying data, understand how decisions are being made and have confidence in the outcomes being generated.
Trust is not a barrier to adoption. It is the prerequisite.
Looking ahead
A final observation stayed with me after the panel concluded.
The fixed income industry has always adopted technology to make markets more efficient, transparent and accessible. AI represents the next chapter of that evolution. But its greatest impact may not be replacing judgement. It may be making judgement more valuable than ever.
As information becomes more abundant and more accessible, the differentiator is unlikely to be who has the most data or even the most automation. It will be who can best combine technology, experience and context to make better decisions.
The winners will not simply be the firms that automate the most. They will be the firms that pair powerful technology with the insight, judgement and trust that remain distinctly human. That, more than any prediction about autonomous trading, was the strongest conclusion I took away from the panel discussion.