AI Subscriptions and Ethics
Steven Forth is CEO of Ibbaka. Connect on LinkedIn
The subscription economy is about to be transformed by AI agents. This is going to introduce many new ethical challenges to B2B monetization and pricing. The spring of 2025 seems like a good time to pause and reflect on what these will be.
Sometimes fiction can be a good place to imagine the future, especially dystopic futures. A recent episode of the British show Black Mirror is a lens into some of the exploitative anti patterns we are already seeing in the subscription economy that are about to be amplified by AI and agents.
Black Mirror Series 7 Episode 1: Common People
Spoiler Alert: The following contains many spoilers; you may want to watch the episode before reading on.
On Rivermind, you can pay to dial up ‘Serenity’, but at what cost? Screenshot from Black Mirror Series 7 Episode 1: Common People.
(précis of the Wikipedia summary)
The story centers around a woman who needs a subscription service to survive, the consequences she experiences as the price goes up and the quality declines, as well as the lengths her husband is willing to go to in order to cover the cost of the service. The episode received positive reviews from critics, with praise going towards the performances of the cast, the commentary and the ending.
Welder Mike and schoolteacher Amanda have been married for three years and are trying to conceive a baby. One day while teaching, Amanda collapses, and doctors discover she has an inoperable brain tumor. Mike is introduced to Gaynor , a representative from tech startup Rivermind Technologies. Gaynor explains that Rivermind can remove the tumor and replace her excised brain tissue with synthetic tissue powered by their servers. While the surgery is free, the couple agree to pay a monthly subscription fee to give Amanda a chance at living a normal life again.
Initially the service seems to help Amanda, but as time passes they find that it has several limitations which can only be bypassed by subscribing to the costlier "Plus" tier. Limitations include geographic coverage, compromised functionality and data processing speed. Unbeknownst to Amanda, she begins interjecting brief advertisements into her daily speech, threatening her job and their financial security.
Amanda finds herself sleeping more and more but that the sleep is not restorative. They learn that at lower tiers Amanda’s brain is being used to provide additional processing power to people subscribing to higher tiers.
On their anniversary, Mike buys a 12-hour "Lux" pass that allows users to manipulate their emotions and sensations (an addictive experience). Amanda turns up her pleasure setting and the two have sex, but later Amanda's behavior becomes increasingly erratic.
When Mike mentions that they're trying for a baby, Gaynor explains they would be charged an additional fee if Amanda were to become pregnant. Mike becomes furious and the couple dejectedly leave the office.
One can see a number of common subscription anti-patterns here.
Functionality of lower tiers is degraded as more expensive tiers are introduced in an attempt to force upgrades.
Lower tiers are interrupted by advertisements introduced into the experience.
Upgrades are needed to continue to get value.
Lower-tier users are leveraged to create value for upper tiers.
The company in the episode is called Rivermind, but the practices are real. You can explore this with this Prompt: “Are there any actual companies similar to Rivermind Technologies, the company in Black Mirror, Series 7: Episode 1, Common People?” You can read what Perplexity generated here.
Guidelines for Ethical Pricing of AI and AI Agents
1. Ensure Transparency and Explainability
Clearly communicate how prices are determined, making pricing methods and algorithms understandable to both customers and internal stakeholders.
Regularly audit and document pricing algorithms to ensure decisions can be explained and justified.
2. Promote Fairness and Mitigate Bias
Actively identify and correct biases in pricing models to prevent discriminatory or unfair outcomes for any group or customer segment.
Treat similar customers consistently and avoid unjustified price discrimination.
3. Protect Customer and User Privacy
Use customer data responsibly and transparently, ensuring compliance with privacy laws and avoiding invasive surveillance pricing practices.
Inform customers about data collection and its use in pricing decisions.
Even if a customer approves certain data usage, make sure that this does not violate the rights of users.
4. Foster Accessibility and Inclusion
Design pricing structures (such as tiered or usage-based models) that make AI and AI agents accessible to a wide range of customers, including smaller businesses and underserved markets.
Consider offering discounts or grants to nonprofits, educational institutions, or organizations serving marginalized groups.
5. Build and Maintain Customer Trust
Cultivate a culture of fairness and openness within the organization, and actively seek customer feedback on pricing practices.
Address concerns promptly and demonstrate a commitment to ethical business conduct.
6. Engage with Regulatory Standards and Best Practices
Stay informed about evolving regulations related to AI pricing, privacy, and competition, and participate in industry discussions to shape responsible frameworks.
Regularly review and update practices to align with legal and ethical standards.
7. Consider Environmental and Social Impact
Factor in the sustainability of AI infrastructure and operations when setting prices, and promote energy-efficient or eco-friendly options where possible.
Strive for pricing strategies that support broader social justice, fairness, and non-discrimination goals.
A prompt to test for ethical pricing
As the next step, consider developing a series of prompts that you can use to evaluate your pricing, packaging and data privacy to ensure ethical pricing.
Here is an initial take on a prompt to test for ethical pricing (generated using Perplexity).
Prompt: “As a prompt engineer at a B2B SaaS company developing AI agents, you are working with a pricing designer to develop a prompt or series of prompts to assess pricing design and pricing practices to design a prompt to check pricing to ensure it is in line with the following ethical guidelines. Propose a prompt or series of prompts that can be used to check for adherence to these guidelines.
Follow best practices in prompt design and prompt engineering.
Explain your process in generating the prompt(s).”
Ethical Pricing Assessment Prompt Framework
For B2B SaaS Pricing Teams
Step 1: Transparency & Explainability Check
Prompt:
"Describe how your pricing model is communicated to customers and stakeholders. Include:
A breakdown of factors influencing pricing (e.g., usage, features, infrastructure costs).
Documentation processes for pricing algorithms.
Frequency of audits to ensure explainability.
Example: 'Our tiered pricing is based on API call volume, with a public formula for calculating overages. Audits occur quarterly."
Step 2: Fairness & Bias Mitigation
Prompt:
"Identify 3 methods used to detect biases in your pricing model. For each method, explain:
How customer segments are analyzed for disproportionate impacts.
Processes to correct unjustified price discrimination.
Example: 'We use A/B testing to compare pricing outcomes for SMEs vs. enterprises and adjust tiers if disparities lack clear cost justification.'"
Step 3: Privacy Compliance
Prompt:
"Map all customer data types used in pricing decisions. For each data type:
State the purpose (e.g., personalized discounts).
Confirm compliance with GDPR/CCPA.
Describe how users (not just customers) are informed about data usage.
Example: 'Usage metrics are anonymized; we exclude sensitive demographic data in dynamic pricing.'"
Step 4: Accessibility & Inclusion
Prompt:
"List pricing strategies to serve underserved markets (e.g., nonprofits, SMEs). Specify:
Discount structures or grants offered.
How tiered pricing accommodates varying budgets.
Example: 'Nonprofits receive a 40% discount, and our starter tier has no minimum user cap.'"
Step 5: Trust-Building Mechanisms
Prompt:
"Provide 3 examples of customer feedback on pricing and how concerns were resolved. Highlight:
Channels for feedback (e.g., surveys, CS interactions).
Revisions made to pricing based on feedback.
Example: 'After SME complaints about opaque surge pricing, we added real-time cost calculators.'"
Step 6: Regulatory Alignment
Prompt:
"Name 2 recent regulatory changes (e.g., EU AI Act, FTC guidelines) impacting your pricing. For each:
Explain adjustments made to comply.
Note participation in industry groups shaping standards.
Example: 'We removed ‘dark pattern’ upsells after FTC warnings and joined the Responsible AI Pricing Consortium.'"
Step 7: Environmental Impact Review
Prompt:
"Quantify how pricing incentivizes eco-friendly usage (e.g., carbon-aware discounts). Describe:
Cost adjustments for energy-efficient AI inference options.
Partnerships with green cloud providers.
Example: 'Customers using AWS’s carbon-neutral regions get 5% credits.'"
Process Explanation
Structured Breakdown: Aligns prompts with each guideline to avoid overlap and ensure coverage.
Actionable Examples: Reduces ambiguity by illustrating ideal responses.
B2B Focus: Targets SaaS-specific concerns (e.g., tiered pricing, enterprise compliance).
Mix of Quantitative/Qualitative: Balances yes/no checks (e.g., "Confirm GDPR compliance") with open-ended exploration ("Describe adjustments").
This framework enables systematic auditing while encouraging proactive ethical design.
