AI Ethics Consulting
for SMEs
AI tools can introduce bias in hiring, scoring, and decision-making before anyone notices. Prompt & Pulse helps you detect those risks, put practical guardrails in place, and meet GDPR and EU AI Act obligations — without the jargon.
- Bias review of hiring tools, prompts, and AI-assisted workflows
- Governance documentation for GDPR & EU AI Act readiness
- Workshops and action plans your team can actually use
Why now
- Bias starts before the model — in criteria, prompts, and workflows
- The EU AI Act applies progressively: major rules from 2 August 2026
- A preventive review costs far less than a GDPR dispute or a reputational crisis
- Ethical AI and performance are compatible — if you build it in from the start
New to AI bias? Read our latest articles →
Choose the right starting point
Three service areas — each with a dedicated page and clear deliverables. Not sure where to start? The ethical AI consulting page explains the full approach.
Ethical AI diagnosis
Map your AI uses, identify priority bias and governance risks, and get a clear written action plan. The natural first step before any deeper engagement.
Explore the diagnosisAlgorithmic bias workshop
Hands-on sessions for HR, marketing, product, or leadership. Real cases from your industry, practical spotting exercises, and materials your team keeps.
View the workshopAI support for businesses
Prompt review, governance documentation, and accountability frameworks for organisations building lasting AI practices. Includes CNIL and NIST RMF alignment.
See full supportWhat to expect after you reach out
Free scoping call — 30 min
We discuss your context: which AI tools you use, which decisions they influence, and what your current concerns are. No preparation needed on your side.
No commitmentEthical review & risk map
We audit your relevant tools, prompts, or processes. You receive a written report with prioritised findings, risk levels under GDPR and the EU AI Act, and practical recommendations.
Written reportAction plan & optional follow-up
Findings become a concrete action plan. Optional follow-up can include rewritten prompts, governance documentation, a team workshop, or ongoing advisory.
Clear next stepsWhy algorithmic bias puts your organisation at risk
Algorithmic bias is not a technical glitch. It reflects — and magnifies — human judgments, intentional or not. Left unchecked, AI can repeat and spread discrimination at scale. Read more in our article on responsible AI for SMEs →
Real, documented cases
Discriminatory hiring tools. Recidivism prediction systems that disadvantage certain groups. Facial recognition with unequal error rates. These are not hypotheticals — they are studied by NIST and cited in EU regulatory guidance.
Impact on real people
Automated decisions affect access to jobs, credit, and services. When systems are opaque, it becomes harder to challenge outcomes. Errors can repeat silently, quickly, and at scale.
A regulatory calendar already running
GDPR already creates duties around automated decisions. The EU AI Act adds layered obligations: prohibited practices from February 2025, GPAI rules from August 2025, full high-risk applicability from August 2026. See the official EU AI Act timeline →
Trust is fragile, reputation is slow to rebuild
An ethical failure in AI can damage trust with clients, partners, and employees before anyone understands what went wrong. Getting ahead of these risks shows your organisation takes responsibility seriously.
Grounded in practice,
not just principles
Prompt & Pulse is founded by Dieneba LESDEMA — an AI ethics consultant with over 25 years of international corporate experience across France, the UK, Zimbabwe, and South Africa.
Her approach is sociotechnical. Bias enters AI systems before training, during training, and after deployment. The relevant questions are not only technical. They concern who defines the targets, who labels the data, who bears the cost of errors, and who can contest decisions.
View LinkedIn profile →What this support is — and is not
Included
- Bias risk mapping
- Prompt review & rewriting
- Governance documentation
- Team workshops
- GDPR & EU AI Act framing
Not included
- Tool selection or IT deployment
- AI model architecture
- Legal counsel or formal audit
- Technical development
Prompt Engineering Certification
Jedha Bootcamp, 2025 — structured methodology for prompt design, bias testing, and safe AI instruction.
25+ years international experience
France, UK, Zimbabwe, South Africa — corporate roles across pharma, education, travel, and international development, including Sanofi, Baxter, and Aga Khan Academies.
Member: Hub France IA & SheLeadsAI
Active in the French and international AI ecosystem, contributing to responsible AI adoption for SMEs and organisations.
Author
Office manager, Assistant, secrétaire… à l'assistant 4.0 — published 2022, based on six months of documentary research on AI's impact on professional roles.
The frameworks we build on
Ethical AI consulting does not exist in a vacuum. These are the regulatory and methodological frameworks that inform our approach.
EU AI Act — official timeline
Phased implementation from August 2024 to August 2027. Covering prohibited practices, GPAI obligations, and high-risk system rules.
View the official timeline →CNIL — practical AI & GDPR guidance
The French data protection authority's practical sheets on AI uses, automated decisions, and GDPR watch points.
Read CNIL AI resources →NIST AI Risk Management Framework
A structured, internationally recognised approach to mapping, measuring, and managing AI risks across the development and deployment lifecycle.
Explore the NIST AI RMF →Built for teams that want clear judgement around AI
SMEs & mid-sized companies
Already using AI, or teams have started — but without a shared framework yet. You want to do it right before it becomes a problem.
HR & recruitment teams
You want to review screening tools, prompts, and candidate-related processes before bias shapes decisions quietly.
Marketing, product & innovation teams
You use AI for writing, analysis, ideation, or workflows — and need clearer validation habits and accountability practices.
Public bodies & non-profits
Committed to fairness, transparency, and setting an example in how you deploy digital tools that affect people.
Not sure which service fits your situation?
Start with the ethical AI consulting pageQuestions people ask before getting in touch
What does Prompt & Pulse help teams with?
Prompt & Pulse helps SMEs and teams review their AI uses, detect algorithmic bias in hiring tools and prompts, train staff on responsible AI practices, and build governance documentation for GDPR and EU AI Act readiness. Engagements produce written deliverables — not just advice.
Is the EU AI Act already relevant for my organisation?
Yes. The EU AI Act entered into force on 1 August 2024. Prohibited AI practices applied from 2 February 2025. GPAI obligations started on 2 August 2025. The majority of rules — including those for high-risk uses such as hiring, scoring, and access to essential services — apply from 2 August 2026, with full roll-out by 2 August 2027. See the official EU AI Act Service Desk timeline →
Where should we start?
Start with the ethical AI consulting page → It outlines the full diagnostic approach: mapping your AI uses, identifying the main risk areas, and prioritising the next practical steps before any deeper engagement.
Do you work in English?
Yes. All engagements — assessments, workshops, written deliverables — are available in both English and French. Sessions run fully online or on-site, with organisations based in France, Europe, or internationally.
What is the difference between AI ethics and regulatory compliance?
Compliance covers legal obligations: GDPR requirements on automated decisions, and EU AI Act duties for high-risk uses. Ethics goes further — it examines real-world impact on people, how decisions are explained, and whether outcomes are fair across groups. Both are addressed in Prompt & Pulse engagements. Neither replaces formal legal counsel.
How long does an ethical AI review take?
A first diagnostic typically runs one to two weeks, depending on the number of tools reviewed. It produces a written report with prioritised findings. A deeper engagement — workshops, prompt rewriting, governance documentation — runs over four to eight weeks.
Ready to take a clear look at your AI risks?
A free 30-minute call to understand your context and identify the most practical first step.
- No preparation needed
- No technical knowledge required
- No commitment after the call
2025 Prompt & Pulse. All rights reserved