Is Your ETRM or CTRM Holding You Back? How AI Is Reshaping

Many energy and commodity businesses still run trading and risk systems implemented ten or fifteen years ago. They work, broadly, but they are often heavily customised, surrounded by spreadsheets and expensive to change. Meanwhile, the platforms on the market have moved on, and artificial intelligence is now one of the main reasons. If you own an older ETRM or CTRM, this is a practical guide to what has changed and what it could mean for you.

Why now?

AI in trading software is no longer a slide in a vendor roadmap. Industry analysts at CTRM Center report that AI features in a large share of current vendor conversations, and Gartner has forecast that around 40% of enterprise applications will embed task-specific AI agents by the end of 2026, up from fewer than 5% in 2025. Trade publications have also noted that energy firms are revisiting their CTRM systems as the technology advances. The question for owners of older systems is no longer whether AI will reach their workflows, but whether their platform and data are ready for it.

What AI is doing inside trading platforms today

1. Turning documents and emails into trades

One of the most mature uses is extracting trade details from confirmations, broker statements, emails and contracts, and converting that unstructured information into structured CTRM data. Instead of operations staff re-keying deals, AI proposes the booking and the system validates it.

2. Conversational trade capture

Some platforms now let a user describe a trade in plain language while AI populates the trade ticket for review and approval. The CTRM still applies its normal validation, permissions and controls before anything is booked.

3. Asking questions of your positions

Natural-language querying lets traders, risk managers and executives ask questions such as "what is our net exposure to Q1 German baseload?" and receive an answer or report without waiting for an analyst. Several vendors now connect AI assistants directly to CTRM data through open standards such as the Model Context Protocol, so positions can be interrogated from familiar tools.

4. Matching, reconciliation and exception handling

AI is being applied to match deals, confirmations, shipments and invoices, and to flag reconciliation breaks. Operations teams spend their time on genuine exceptions rather than routine checks.

5. Smarter risk monitoring

Machine-learning models can analyse historical patterns, counterparty behaviour and market conditions to spot emerging credit or margin issues earlier, and to highlight unusual activity for review.

6. Agents for lower-risk operational tasks

Vendors are working towards AI agents that complete lower-risk operational activities on their own, such as preparing reports, chasing missing data or drafting settlement documents, within defined boundaries. This is promising, but it is as much a governance question as a technology one.

7. AI-native platforms

New entrants are building CTRM products with AI designed in from the start, while established vendors are adding AI to existing products. This widens the field of credible options for firms considering replacement.

The benefits for trading companies

Lower operating cost: less manual entry and reconciliation, and faster month-end close.
Fewer errors: automated extraction and matching reduce re-keying mistakes and the risk they carry.
Faster decisions: traders and managers get answers in seconds rather than waiting for reports.
Better risk control: earlier warnings on credit, margin and anomalies.
Scalability: volumes and new markets can grow without matching growth in back-office headcount.
Better use of expertise: experienced staff focus on judgement and exceptions, not routine processing.

Why older systems struggle to benefit

AI is only as good as the data and architecture beneath it. Analysts have observed that the real dividing line will not be which vendors can claim AI features, since almost all soon will, but which platforms can use AI safely and effectively. Systems held back by poor data quality, ageing architecture, heavy customisation or weak implementation will struggle. Common symptoms in older environments include:

Key data held in spreadsheets outside the system of record
Customisations that make upgrades risky and expensive
Limited APIs, making it hard to connect AI tools or new data sources
Inconsistent reference data across trading, risk and finance

Governance cannot be an afterthought

European regulators have highlighted explainability, human oversight and reliance on third-party AI providers as key risks of AI adoption in financial markets. For trading firms, that translates into clear rules about what AI may propose versus what it may execute, full audit trails, and controls that keep a human accountable for every booking and limit decision.

Upgrade, extend or replace?

There is no single right answer. Some firms can add AI capabilities around an existing platform through modern integration and data layers. Others will gain more from upgrading to a current version of their platform. For some, particularly those with heavily customised legacy systems, a replacement is the most cost-effective route. A structured approach helps:

Assess current processes, data quality and the cost of manual work.
Prioritise the use cases with the clearest return, often trade capture, reconciliation and reporting.
Fix the data foundation so AI has reliable inputs.
Evaluate options independently, including AI capabilities, governance features and APIs, through scripted demonstrations using your own scenarios.
Pilot before scaling, with clear success measures.

Orivyn helps firms through each step, from an independent as-is assessment to vendor selection and implementation. See also our guide to running an ETRM or CTRM vendor selection.

Sources

Market figures are as reported by the sources above at the time of writing and will change; they are provided for context, not as investment advice.

Is Your ETRM or CTRM Holding You Back? How AI Is Reshaping Trading Platforms