AI principles
These principles define when we create AI experiences for customers, what role AI should play, and how it shows up responsibly in our products.
Foster trust with transparency
As designers, we hold a critical responsibility in shaping a customer’s relationship with AI. AI systems can often feel mysterious, leaving customers confused about how responses were generated or why a decision was made. This lack of visibility can lead to confusion, frustration, and ultimately, a breakdown of trust.
At Intuit, we solve for this by making information accessible without increasing cognitive load. Trust is rooted in clarity. Always be clear about when the customer is talking to a bot, and avoid overselling what AI can do for customers. Setting the right expectations will impress customers more than overpromising and underdelivering on what AI can actually do.
See also: Trust, transparency, and control, Mentioning AI
Here's an example where the customer is chatting with AI, and we provide a data point paired with a clear source. In product, we'd also have our AI disclaimer on this screen to give customers access to information about how we use AI.
Focus on the customer benefit, not the technology
Our primary goal is to communicate the tangible benefits and seamless functionality of our products to our customers. Overemphasizing the technology itself can distract from the customer’s goals.
Focus on the customer and the jobs AI is getting done for them. The design patterns should make it clear that AI has taken the action. The content should focus on the work, not what we're doing in the background to make it happen.
See also: Mentioning AI
94% of your expenses were categorized. Review the remaining 6% to finish your books.
Our AI-powered transaction classifier used machine learning to categorize 94% of your expenses automatically.
Aim to reduce cognitive load
AI experiences should reduce cognitive load, not add to it. Avoid making customers read long text blocks, answer too many questions, or make complex decisions. AI should provide a clear point of view and a suggested course of action based on its analysis.
Something to note: AI isn’t a salesperson, so it shouldn’t push products. When presenting a solution, a product or subscription can be options, but those should also be paired with an alternate path that doesn’t cost the customer extra money.
Here’s the top-performing option for your audience: “Don’t miss your exclusive offer.”
Want more ideas?
Choose from the following 12 subject lines ranked by predictive engagement score, semantic variation index, and audience segmentation heatmap.
Be brief and clear
Text-based AI experiences are usually in space-constrained areas with limited formatting. In a chat experience, long responses force the customer to scroll, making it harder to read. In a tile or product suggestion, there are only so many characters.
Voice-based bots or avatars put the burden on customers to remember what was said. But people can only remember so much at a time. For any AI experience, be sure to prompt and evaluate for short, organized responses. A customer can always ask the bot for more info if needed. For more on this, see our guidance on prompting and evaluation.
Your average click rate is down 8%. You could try scheduling this email campaign at 6pm instead of 10am. Would you like me to make that change?
Your last five campaigns showed a downward trend in click-through rates, particularly in your segmented loyalty audience, which might indicate fatigue with your current promotional messaging strategy...
Conversation best practices
Effective conversation design helps AI interactions feel purposeful rather than robotic.
Good conversational AI:
- Helps customers reach their goal with minimal effort.
- Responds to context across the entire conversation, not just individual turns.
- Respects customer’s time, attention, and emotional state.
- Anticipates unspoken needs.
- Lightens customers’ cognitive load by leading with a point of view.
Conversational maxims for AI
In the 1960s, language philosopher Paul Grice introduced a cooperative principle to explain how people communicate effectively. He identified four key categories, known as Gricean maxims:
- Quantity: Be informative
- Quality: Be truthful
- Relation: Be relevant
- Manner: Be clear
AI experiences that break these norms quickly become frustrating or untrustworthy.
Examples
Some more conversational maxims that AI experiences should follow:
- Goal-oriented. Design conversations around real customer goals, validated through user research—not system capabilities.
- Quick and clear. Customers expect software to be faster and more precise than humans. Be concise, direct, and accurate.
- Turn-based. Avoid long system monologues. Make it clear whose turn it is and what the customer can do next.
- Polite. Acknowledge context and emotional state. Never talk down to customers or imply something should have been obvious.
- Error-tolerant. AI must gracefully handle typos, incomplete input, made-up words, business names, and non-English names. Customers should never feel punished for imprecision.
When to use generated content
Creating generated content requires a bigger upfront investment than static content. When teams are trying to decide what to use, consider the relative returns on these factors:
- Success metrics and business goals
- Performance/quality
- Data availability
- Cost
- Risks
- Time/complexity to implement
Most experiences that involve AI contain a mixture of content types. Based on the above factors, define which elements of a design should use which type of content. Before committing to a generated content solution, be sure to evaluate the alternatives, like static or dynamic content.
- Static: Written once by a human, never changes
- Dynamic: Written by a human, with variables that are replaced (Good morning, George! / Good evening, George!)
- Generated: Written by an LLM