How does the AI in Supplier Data Manager work?

The AI in Supplier Data Manager (SDM) operates in two layers:

Understanding inputs: The first layer interprets your input and translates it into a format the system can process. This layer has been trained on a vast multilingual dataset, enabling it to understand input even in regions where SDM might not yet have active customers.

Producing outputs: The second layer is specialized for specific tasks, converting the processed input into actionable outputs, such as categories or attribute values. This layer can be trained individually for each customer to ensure data security when a trained model is used. Data from one customer is never shared with or accessible to another.

How is the accuracy of the AI ensured?

For trained models in Supplier Data Manager, accuracy is maintained through a robust auditing process:

• Predictions made with high confidence (automated rows) are sampled and flagged for auditing.
• Auditors validate these samples by comparing the AI's output with user-approved corrections, ensuring the system performs reliably.
• The sampling rate is customizable and typically ranges between 10% (for new models requiring trust-building) and 1% (for established models with proven performance).

What data is used to train the AI models?

In Supplier Data Manager, the AI learns from different data sources depending on the model type:

Trained models: The AI learns from user-approved actions:

• Training data: Rows that are manually validated or corrected by users are used to improve the AI's algorithms.
• Excluded data: Automated rows are excluded from training to prevent biases, even though auditing mitigates risks of errors in automated outputs.

LLM models: The AI is trained on large datasets and can be used right away, with no additional training period required.

Can SDM use my existing data for training?

Yes, leveraging your existing data is an integral part of the onboarding process for trained models in Supplier Data Manager. During onboarding:

• Akeneo audits your data to verify its alignment with your target taxonomy, correctness, and suitability for supplier or seller requirements.
• This audit ensures that as much of your data as possible is used to enhance the AI's performance.

Is customer data shared? Can I benefit from others' data?

No, Supplier Data Manager ensures complete data isolation. Each customer has a dedicated AI model, meaning:

• Your data is never shared with other customers.
• You cannot access or leverage insights from other customers' data.

What does "Zero-Shot" or AI Classification mean?

AI Classification in Supplier Data Manager uses a pre-trained LLM (large language model) that provides instant results without any training period. It is used by default at the Classification and Extraction steps. See the AI agent overview.

How does AI Classification understand your data?

AI Classification in Supplier Data Manager uses LLMs trained on extensive sets of text, enabling them to interpret and classify product attributes from the start — no onboarding or training data required. SDM uses high-quality AI models to ensure accurate, real-world results across a wide range of industries and languages.

What if the AI makes a mistake?

AI can make mistakes. LLM models are designed for high accuracy but are not infallible. You can correct data manually during each step in Supplier Data Manager. For the Classification step, you can also guide the AI using custom prompts — see AI agent for details.

Can the AI Classification agent learn from corrections?

No, AI Classification (zero-shot) models in Supplier Data Manager do not rely on feedback loops. Prompt-based customization is available for the Classification step, allowing you to provide specific instructions to refine the AI's behavior. See AI agent for instructions on writing custom prompts.

Can I customize the AI with custom prompts?

Yes, Supplier Data Manager lets you add a custom prompt to the Classification step to guide the AI toward more accurate results for your specific taxonomy and context. Custom prompts are also available for the Extraction step, where you can configure which source attributes the AI uses.

To add a custom prompt to the Classification step:

1. Go to Workflow > Settings > Steps.
2. Select the classification step you want to customize.
3. Enter your instructions in the text box provided.

For tips on writing effective prompts, see AI agent.

Does the SDM Mapping module use AI?

Yes, Supplier Data Manager's Mapping module includes an AI-powered mapping suggestion feature. When you click Generate AI mapping, the AI analyzes column names and sample values from your source file, then proposes the most likely matches for each unmapped field. Every suggestion includes a confidence score — only matches above 70% are surfaced. You can accept, adjust, or ignore each suggestion.

See Modules questions for more details on AI mapping.

Which AI models are used in Supplier Data Manager?

Supplier Data Manager uses AI-powered large language models (LLMs) for different steps:

• Classification (zero-shot model): An LLM-based model for categorizing products into your taxonomy.
• Extraction: An LLM-based model for extracting and enriching product attributes from text.
• Mapping: An LLM-based model for generating field mapping suggestions from your source file.

For details on how these models handle your data, see the security question below.

What about security?

Supplier Data Manager uses enterprise plans with OpenAI and Google, ensuring that data provided for content generation is not used in model training. This data flows exclusively within their servers, returning directly to Akeneo without storage or reuse. Only the precise information needed to generate accurate content is shared with these providers.

Akeneo does not share client metadata (such as company names) with OpenAI or Google. If you choose to include such information in your custom prompts, that is your decision — Akeneo does not add this information on your behalf.