Develop AI Trust & Accountability
Cultivate company-wide culture-based governance.
New rules must govern AI-charged businesses that are lived daily to enable growth. A pervasive governance ethos must infuse corporate culture, foster trust, guide and track dynamically. Multi-level accountability must be distributed and individualised - mindset and behaviours first, policies second.
“Project Glasswing by Anthropic has just identified ‘thousands of high-severity vulnerabilities, many of them critical, in every major operating system and browser, along with a range of other important pieces of software.” Anthropic.com April 2026.
Who’s Responsible for AI?
Governance may have felt less critical when businesses advanced at a slower, steadier pace in familiar, linear directions. Now, AI apps are evolving rapidly, AI agents are making decisions, and serious vulnerabilities are exposed. High risk, unpredictability, and ambiguity at velocity are a given. Governance is crucial:
59% of EU firms aren’t sure how fast they could shut down AI in a crisis.
21% feel confident stopping an AI system within 30 minutes.
Only 20% of workers unsure of who is responsible for AI failures.
Only 42% could explain an AI failure to leadership or regulators.
Only 38% of workers can accurately pinpoint who’s accountable in their business [TechRadar Insider, March 2026].
As agents proliferate, Responsible AI (RAI) is essential, defined by McKinsey as: strategy, risk management, data and technology, governance, and (starting 2026) agentic AI governance and controls. Levels of RAI Maturity are defined:
[McKinsey, State of AI trust in 2026]
“As AI systems take on greater autonomy, AI trust and the responsible AI (RAI) practices that enable trust are no longer a tangential concern but a foundational requirement for realizing the full potential of the technology.” McKinsey, State of AI trust in 2026: Shifting to the agentic era, March 2026.
Responsible AI maturity is improving, but lagging in areas of governance and agentic governance. Security and risks are stalling the growth of agentic AI at many companies, while inaccuracies and cybersecurity continue to be the most frequently cited risks as AI adoption increases.
[McKinsey, State of AI trust in 2026]
Pre-emptive actions by organisations lag behind executive’s awareness of risk across nearly every AI risk category. While incidents involving AI are not increasing, confidence in organisations’ responses has dropped.
Lack of confidence is unsurprising: just 8% of 3,048 Russell 3000 and S&P 500 companies disclose board-level AI oversight (insights.issgovernance.com).
“By 2027, 60% of organizations will fail to realize the anticipated value of their AI use cases due to incohesive ethical governance frameworks” Gartner, ‘Turn Data, Analytics & AI into Strategic Growth Drivers‘ 2025.
Who is accountable for people using AI responsibly at your company?
Grounded Governance
AI Governance outlines how your organisation uses AI to ensure operations are within legal and ethical boundaries that align with agreed values and societal norms. RAI needs are growing. Near-term strategic planning assumptions:
By 2027, AI governance and responsible AI capabilities will be integrated into 75% of AI platforms, making them the main area of AI competition.
By 2027, AI governance will become a requirement of all sovereign AI laws and regulations worldwide [Gartner, Reference Guide for AI Governance].
“Organizations are working to strengthen the foundations of AI trust—closing capability gaps, clarifying accountability, and building the RAI capabilities needed to ensure trust accelerates innovation rather than constrain it.” McKinsey, State of AI trust in 2026: Shifting to the agentic era, March 2026.
System Not Function
Some companies have one senior leader accountable for overall governance. In others, responsibility is divided up or isn’t yet fully clarified. As AI proliferates system-wide, ownership must exist beyond tech functional boundaries:
[PwC, 2025 Responsible AI Survey, Oct 2025]
Accountability must extend throughout organisations’ divisions, levels, and teams. 56% of executives surveyed say their first-line teams—IT, engineering, data, and AI—now lead Responsible AI efforts. PwC advises three lines of defence models—built for speed and trust:
First line: Builds and operates responsibly.
Second line: Reviews and governs.
Third line: Assures and audits [PwC, 2025 Responsible AI Survey].
“Responsible AI is a team sport” [PwC] makes sense as team members use AI to innovate and influence, motivate and monitor each other. However, every employee is first governed by their own rules before they align and collaborate.
When nobody is watching, workers must take personal charge of how they trust and entrust powerful AI applications and agents still in development.
Develop distributed culture-based governance of individual AI accountability. Every employee in their daily work takes some responsibility for cleaning data, checking errors, asking critical questions, tightening limits, adding safety tests, mitigating flaws in AI processes, and paying attention as they evolve.
How AI accountable are you in your daily working practices?
Trust: Guard & Grow
Governance is a trust-building capability. AI trust is critical to scale. Without it, your company is likely to experience reduced AI usage and integration, greater impact from negative situations, and reduced stakeholder trust.
“By 2028, 50% of organizations will implement a zero-trust posture for data governance due to the proliferation of unverified AI-generated data.“ Gartner, Jan 2026.
LLMs ingest and learn from the totality of internet-posted information. The growing volume of approximate or inaccurate AI-generated data is therefore compounding and proliferating, also, at times, obscuring correct information to power the answers we need.
“Organizations can no longer implicitly trust data or assume it was human generated.”“As AI-generated data becomes pervasive and indistinguishable from human-created data, a zero-trust posture establishing authentication and verification measures, is essential to safeguard business and financial outcomes.” Wan Fui Chan, Managing VP at Gartner.
AI trust must be cultivated as a core capability going forward. Trusted AI solutions are “reliable, transparent, fair, resilient, and accountable” [OECD].
[McKinsey, State of AI trust 2026]
AI trust requires a deliberate combination of people, processes, policies, and technologies so that accurate, safe and effective human-AI collaboration can be developed with confidence.
#AIAccountability will not wait.
Involve every individual. Engage each person in the part they play in the process to be able to test the tools, inputs, limits, guardrails, risks, and outcomes.
Cultivate a trust-based culture of individual AI accountability.
RESEARCH for reference and next steps:
News & Muse
📹 Building Effective AI Accountability Frameworks, Dr. Pavan Duggal
📘 Governing the Machine: How to Navigate the Risks of AI and Unlock Its True Potential, Ray Eitel-Porter, Dr Paul Dongha and Miriam Vogel
🗞️ 2026 Cybersecurity Issues, Gartner.
🎶 Accountability, Deraj Global
Every executive and employee is accountable for engaging responsibly in AI utilisation to improve business models, processes, and outcomes.
Each person must step up as complexity increases, and pay attention to the role and responsibilities they give and rely upon AI for. Foster a pervasive thorough top-down-bottom-up-across-the middle approach to governance to enable sustainable growth.
See you next week.
Sophie
If you’re interested to learn more about my work and insights:
modernworkLIFE.co -- My content hub, learning platform, and the place to book private leadership advisory sessions with me. Register to be notified when I add new content; it’s free.
YouTube -- Full versions, shorts, and clips of my interviews with global leaders navigating growth and change.
Substack -- A complete archive of my weekly “Work in Progress Report” newsletter, along with additional content for subscribers.
Podcast -- You can access my “Transforming Work” podcast episodes on Apple, Spotify, and iHeart.
workforce innovator, advisor, speaker, instructor, author
If you have a story, challenge, idea, or insight you would like to share, I’d love to hear it. Just hit reply or connect on LinkedIn.
Scalable strategies. Tactical talk. Workforce transformation.