Artificial intelligence (AI) — it’s everywhere you look and on every business leader’s mind. Many companies are starting to put AI at the forefront of their strategy, but nearly all (89%) CFOs have concerns about their company’s ability to implement AI-led strategies, according to our most recent CFO survey. Given the pace of advancement around the technology, these figures make sense. AI is changing so rapidly that it can be tough to keep up with it all.
Business leaders need to have robust discussions with IT on how to develop a safe and transformative approach to AI. This starts with a clear understanding of the types of AI available and how they can impact the business. This glossary list can help by breaking down complex terms and providing context on how the technology can be used — from day-to-day operations to long-term initiatives.
AI basics
It’s helpful to understand the different types of AI, which can be broken down into four categories:
- Artificial intelligence (AI): Machines perform tasks that typically require human intelligence, such as learning, problem-solving, understanding natural language, and recognizing patterns. Typically, AI usage refers to one single, specific type of task, like organizing a company’s purchases into auto-managed catalogs.
- Machine learning (ML): As a subset of AI, machine learning enables machines to learn from data without being directly programmed. ML algorithms improve performance over time as they are exposed to more data. An example of ML is a purchase order (PO) being automatically matched to an invoice.
- Deep learning: Mimicking the human brain’s structure, deep learning is a specialized form of ML that involves artificial neural networks with multiple layers. It’s effective in processing complex data like images, speech, and natural language.
- Generative AI (Gen AI): An advanced AI technology that focuses on creating data, content, or outputs that are not directly based on existing examples but are generated by the AI system itself. Typically, Gen AI is used for creative or complex tasks, like an AI chat assistant that can answer questions and complete key tasks on a spend management platform.
AI for finance
AI-driven tools for finance teams have the power to automate time-consuming and error-prone tasks like data entry, invoice processing, and expense categorization. More importantly, AI-driven automation gives finance teams the resources and time they need for more strategic initiatives. AI analytics can also empower finance leaders with actionable insights to cut costs, reduce risks, and forecast more accurately.
Invoice data extraction
What it does: It fully automates data extraction from invoices sent from a supplier system and pulls them into your spend management platform. There’s no need to manually enter the data or use separate optical character recognition (OCR) technologies. This drastically reduces transcription errors and boosts accounts payable (AP) team productivity. The data extraction process, like Coupa’s InvoiceSmash, learns from corrections to improve over time and makes the appropriate adjustments where necessary to improve accuracy on future invoices.
Third-party risk management
What it is: Advanced algorithms continuously monitor outside data sources and internal data to help you understand and flag risk across your supplier base. Third-party data — including InfoSec, Anti-Bribery Anti-Corruption (ABAC), GDPR, and other domains — are pulled, analyzed, and aggregated into risk scores. This ensures your company works with ethical and low-risk suppliers.
Fraud detection
What it is: AI-powered fraud detection, like Coupa’s Spend Guard, automatically analyzes spend transactions to uncover hidden patterns or anomalies that could signify potentially fraudulent activity. On top of replacing manual and time-consuming auditing processes, AI-powered algorithms flag and intercept even the most inconspicuous suspicious activity quickly and efficiently — saving businesses money, ensuring compliance, and protecting their reputations.
Spend classification
What it is: Spend classification groups spend data for similar goods and services so that businesses understand their spend and can make more strategic decisions. Manual approaches to spend classification tend to fall apart as businesses grow, leading to siloed, inaccurate data and blind spots in spend visibility. AI-powered spend classification standardizes, classifies, and enriches spend data from across multiple ERP and spend systems, empowering businesses to identify saving opportunities, develop accurate taxonomies, and integrate with external data sources to uncover insights that lead to more strategic, confident spend decisions.
Benchmarking
What it is: Coupa AI aggregates and anonymizes billions of dollars of spend data to help organizations benchmark their performance against peers within or outside their industry, geography, or market. AI generates prescriptive insights for everything from improving ESG objectives to understanding on-contract spend. See the latest Benchmark Report.
AI for procurement
Procurement teams can harness AI to simplify workflows, improve on-contract employee spending, improve visibility into supplier activities, and uncover ways to drive cost efficiencies. In particular, AI-powered spend analysis provides deep insight into spending patterns, contract negotiations, and compliance policies. With improved visibility and automation over the entire procurement lifecycle, teams are freed to focus on high-impact tasks.
Intake and orchestration
What it is: Intake — purchase requests from employees who don’t use procurement systems often — needs to be as simple as possible to drive on-contract spending and reduce compliance risk. Ultimately, streamlining the process of obtaining goods and services helps employees do their jobs and benefits the organization as a whole. An AI-driven intake experience that guides employees through what they need while surfacing preferred items at pre-negotiated prices ensures everyone buys against the latest contract terms. Throughout the buying workflow, automation triggers all the processes needed to fulfill the request, like sending the request up the approval chain or for risk management review. We call this automated orchestration.
Supplier performance and risk
What it is: When it comes to supplier management and risk, businesses can bring supplier information, POs, and invoice data into a single place, then leverage AI to monitor outside data sources, cross-community behavior, and user-submitted performance feedback to understand their supplier base. With Coupa, you can leverage a community of more than 10 million suppliers, plus insights into supplier ratings and behavior. AI can quickly identify top-rated, low-risk suppliers and even help organizations meet ESG and DEI goals by prioritizing suppliers that align with environmental and diversity values.
Spend monitoring
What it is: AI enables comprehensive coverage and real-time monitoring across all spend — reducing a business’s reliance on manual processes or auditing services. AI/ML technology can analyze transactions in real time for suspicious activity or user error with greater accuracy than traditional rule-based algorithms, whether at the supplier or employee level.
Sourcing activities
What it is: AI empowers more strategic decision-making across the entire contract lifecycle. AI-driven insights enable higher-quality catalogs and punchouts that guide users toward on-contract goods and services, ensuring that your business is consistently taking advantage of negotiated contracts and prices. AI can auto-manage catalogs and provide prescriptive insights to identify improvement opportunities.
“The time to embrace AI is now. In today’s dynamic market, those who leverage technology to gain foresight and optimize operations will stay ahead. It’s not about replacing people with AI, but augmenting their capabilities and evolving alongside the technology.”
AI for supply chain
AI can simplify and streamline the complex nuances of supply chain operations, allowing companies to optimize shipping and logistics, maximize distribution center placement, streamline demand forecasting, and more. In particular, the ability to turn a physical supply chain into a digital model creates a risk-free testing environment and empowers companies to develop alternative plans and safeguard their top line when market conditions change — which is happening far more often than ever.
Supply chain modeler
What it is: Pressure test your supply chain using advanced AI/ML technologies and a digital twin — a digital replication of a physical supply chain. Use AI to model real-world scenarios, uncover trends, and plan for supply chain disruptions before they happen. Automatically connect, cleanse, and harmonize supply chain data, such as logistic rates and carbon emissions. Then, leverage prescriptive insights and powerful algorithmic engines to optimize supply networks, inventory, cost-to-serve, and transportation in various use cases.
Demand modeler
What it is: Pairing advanced AI algorithms with smart use of internal and external data has the power to bring new context to demand across any organization. ML can clearly and accurately pinpoint demand signals when applied to historical demand models to help organizations find the best data-driven models for their business. AI can tie in external data such as macroeconomics and world events to understand causal factors, test new demand scenarios, and enable faster, smarter decisions, even as conditions rapidly change.
Supply chain prescriptions
What it is: Prescriptive, AI-powered insights help businesses optimize their supply chains to meet ESG objectives. Community AI data and insights help organizations benchmark against industry peers to understand areas of strength and opportunities for improvement. AI can quickly identify and prioritize suppliers that align with ESG objectives to help businesses meet stakeholder expectations, stay compliant, and manage reputational risk.
Generative AI for all departments
We’re now entering the generative AI era, where language-learning models could automate up to 70% of repetitive and time-consuming work for highly skilled workers. This productivity boost has the potential to free your best high-level employees to focus on strategic initiatives, reinvest time into training junior employees, and support more sustainable growth as the business scales without adding additional workloads.
One of Coupa’s most recent AI innovations, the Coupa Navi™ AI agent, focuses on unlocking the power of generative AI for companies of all sizes. It will provide real-time support to help team members make smarter decisions faster by guiding intake and purchase requests, auto-populating forms, creating personalized dashboards, surfacing key data and insights, and speeding up approval cycle times.
Coupa’s approach to AI
It’s important to remember that any AI — and its potential to help businesses drive smarter and more profitable decisions — is only as good as the data used to train it. Coupa’s AI data is informed by $6 trillion of global transactions from a massive network of over 10 million customers and suppliers, all designed to help businesses free up newfound capital to fund durable, profitable growth strategies.
Our journey to safe, ethically sourced AI started more than 15 years ago. Now, more than 70 AI features are embedded in our AI-driven Total Spend Management platform, and we aim to keep pushing innovation. At Coupa, we take a six-core principle approach to using AI.
Accountability:
- We actively monitor and incorporate findings from industry-leading developments and regulatory changes like the Organization for Economic Cooperation and Development (OECD) AI recommendations, the European Union AI Act, and Executive Order 13960 to ensure compliance with all laws and regulations.
- We regularly seek feedback from our users to ensure our customers benefit from AI in a compliant manner.
Security:
- We conduct regular code reviews, use the latest encryption techniques, and ensure robust authentication and authorization mechanisms by using tools like multi-factor authentication to prevent unauthorized access.
- Our security and privacy programs are subject to regular third-party audits, and our customers can access our compliance reports and certificates.
Safety, fairness, and equity:
- A cross-functional oversight board reviews every AI feature before releasing it to prevent bias. This includes conducting thorough testing and validation to ensure that the output of AI solutions does not compromise the health, safety, or fundamental rights of our users.
Transparency:
- We provide comprehensive product documentation — available through the Coupa Compass portal — explaining how our AI capabilities work, disclosing the data sources used in training the AI system, and providing information about the training process itself.
- We never train public AI models on customer data or process it on unauthorized third-party services.
Human oversight and monitoring:
- Our AI-driven technology incorporates mechanisms that enable “human-in-the-loop” steps when AI is making decisions. This includes models that provide alerts or notifications for human review.
- Our AI addresses common spend management operational pain points and enhances employees’ performance, not replace them.
Validity and robustness:
- We regularly test our AI tools for validity — ensuring they produce results that are aligned with the intended purposes and objectives.
- We regularly test our AI tools for robustness — ensuring our AI systems maintain consistent and accurate performance. Coupa AI can be trusted to deliver reliable and meaningful results.
Embracing the future of AI
It’s apparent AI has the potential to transform how businesses operate, but leaders should also think carefully before adopting the technology. From security concerns to technical questions to process updates, there’s a lot to discuss before rolling out AI across a company.
But the reality is that most finance leaders are not having these discussions at all. Three out of four say they are never aligned with the CIO on the company’s strategic priorities. Taking the time to collaborate with all departments — particularly IT — on a comprehensive AI strategy could drive true operational transformation.
