Chemical companies have spent heavily on AI for laboratories and production lines, but their commercial teams remain underserved. Speciality Chemicals Magazine reports that the next wave of value lies in giving technical salespeople governed tools that answer complex customer questions with speed and accuracy-not just plausible-sounding responses.
The gap is costly. A small group of specialists typically supports a much larger customer base spread across different formulations, regulations, and application conditions. When a cosmetics formulator asks whether a substitute preservative still meets a revised EU Cosmetics Regulation annex at a specific use level and pH, the answer cannot come from a single database. It often requires pulling data from a product information management system, a CRM platform, regulatory records, and a long email chain to a subject expert.
Dr Farid Mirmohseni, chief executive of Kimia, argues that digital investment has largely bypassed the commercial function. "The industry's digital investments have largely ignored the commercial and technical sales function, even though that is where scalable expertise could unlock growth," the article states, summarising his view. Customers are asking more technical questions in real time as reformulation pressure and supply volatility reshape buying decisions.
Why a polished demo falls apart under commercial pressure
Many companies layer generative AI onto existing workflows and expect transformation. Pilots work in controlled settings, then break down when real customers, multiple constraints, and live commercial pressure enter the picture. The tool may find the right document, but technical sales needs more than retrieval. It must judge which source matters when documents conflict, whether a claim is robust enough to support a recommendation, and how a change in humidity, regulation, or formulation could alter the answer.
McKinsey has estimated that chemical companies redesigning their operating model for digital tools and AI can deliver 10 to 20 per cent above-market growth. That upside explains the intensifying pressure to adopt AI, but the magazine suggests many leaders still think in terms of deploying software rather than redesigning the decision-making process around it.
The three layers that determine whether AI works
The article argues that companies should focus less on chasing the most advanced model and more on how the system behaves inside the business. Three layers matter most. At the data layer, the question is what the system can access-safety data sheets, regulatory databases, controlled sources-and where access must be restricted. The platform layer covers access permissions, source ranking, validation, and how uncertainty is handled. The application layer is where business impact appears: can the tool surface the right information quickly, recommend an approved alternative, and help the team act rather than just search?
Testing must mirror the questions commercial teams actually receive. A polished demo is not enough. The real test is whether the platform handles regulatory shifts, changing product constraints, and judgment calls that do not fit a template. For sales professionals in technical industries, this distinction is critical. A learning path like AI for Technical Sales Representatives addresses exactly this gap-moving beyond generic demos to tools that support real commercial decisions.
Governance and human oversight are not optional
The governance challenge is already visible. MIQ's 2025 findings, cited in the article, showed that employees at more than 90 per cent of companies were already using personal AI tools for technical work, often without formal oversight. In regulated industries like chemicals, a poorly grounded answer can quickly become a customer, compliance, or reputation problem.
Human oversight remains essential. The article stresses that AI cannot function commercially without people embedded in the process. That means capturing not only structured data but also the tacit knowledge sitting with experienced staff. One example underlines the difficulty: a customer reportedly needed 18 months to manually compile more than 30,000 knowledge points, revealing how much expertise remains trapped in fragmented formats and informal workflows.
Human verification is a continual process, not a one-time control. A technical expert must validate sources, monitor drift, and update the system as underlying data changes. The aim is not to replace judgment but to make it accessible at scale.
Why this matters for sales professionals
The commercial case for AI in chemicals is shifting from efficiency alone to revenue protection and growth enablement. If a sales team answers technical questions faster, more consistently, and with clearer confidence, it reduces delays in quoting, improves customer trust, and supports reformulation decisions in a volatile market. That is particularly relevant for those selling into global supply chains where regulatory detail and formulation performance determine whether a customer stays or switches supplier.
AI that only summarises information is not enough. The advantage belongs to companies that turn scattered knowledge into governed, usable judgment. For sales professionals, the message is direct: the tools that matter most are not the ones with the smartest model, but the ones built around trust, context, and human accountability. Investing time in understanding how AI for Sales works at this practical level-rather than chasing novelty-will separate teams that scale their expertise from those that remain constrained by it.
Your membership also unlocks:
