Category Archives: MyWeb

SAP Told Its Customers Which AI to Use. Three Per Cent Listened.

Posted on by

By February 2026, DSAG’s own Investment Report had already surfaced the gap SAP would spend the rest of the year trying to close by other means. Only 3 per cent of surveyed companies ran their production AI use cases on SAP’s own tools, while 77 per cent ran theirs on non-SAP solutions instead, Microsoft Copilot chief among them. Two months later, SAP rewrote its API policy. Section 2.2.2 of the new terms blocks external AI agents from calling SAP systems directly, prohibiting what it calls independent scheduling or execution of API calls. Every agentic use case now has to route through Joule, SAP’s own assistant, adding a second layer of inference and cost to anything a customer’s AI tools want to do inside SAP.

 

Two Competitors Bet the Opposite Way

Salesforce and ServiceNow read the same gap differently. Salesforce launched Headless 360 in April 2026, giving external agents direct access through REST, Model Context Protocol tools and the command line, no mandatory detour through a proprietary assistant required. ServiceNow’s Action Fabric, announced at Knowledge 2026, opened two decades of workflows and business rules to any agent through the same open standards. Neither company built a legal wall around its intelligence layer. Both are betting that customers will choose their AI because it is good, not because the contract requires it.

 

The Mandate Had a Structural Problem Too

Joule is only available to customers on RISE or GROW with SAP cloud contracts, which shuts out the thousands of organisations still running on-premise ECC. Support for ECC 6.0 ends in 2027, so over ten thousand customers are being asked to migrate their core system and adopt a mandated AI layer on the same timeline, set by the vendor rather than the customer. SAP’s own chief customer officer has disputed the DSAG survey’s reach, noting that fewer than 100 organisations responded, which is a fair challenge to the precision of the number. It does not explain why SAP is separately funding a hundred-million-euro partner investment programme to accelerate adoption of a product it says the market already wants.

 

The Hedge That Followed the Mandate

At Sapphire 2026, a few months after the API policy took effect, SAP quietly hedged its own bet. Alongside Joule Studio 2.0, it introduced an AI Agent Hub, described as vendor-agnostic governance for SAP and third-party agents alike, bundled into the platform at no extra charge. A company confident that mandating Joule had closed the gap DSAG measured in February would not need to build the open door it spent the spring telling customers they did not need.

 

The Question Every PMO Should Be Asking Now

None of this is really about SAP. Every PMO running a platform strategy right now is making the same call in miniature, whether to route AI agents through one sanctioned tool or leave the door open for whichever tool the business actually adopts. SAP had the market power to write the mandate into a contract, and it wrote that mandate only after its own user group had already published the adoption numbers it was meant to fix. Assume your organisation has less leverage than SAP did, and plan the exit clause before you write the mandate, not after the adoption numbers come back.

Patience Is Not Passive. It Is the Hardest Active Choice Most People Never Make

Posted on by

Patience gets treated as the absence of action: the thing you’re doing while you wait for something else to happen. The clearest evidence says that’s backwards, not from child-development research, but from what happens when trained professionals are given a straight choice between waiting and acting. Patience is usually the harder, more active choice, and impulsive action is often the easier one dressed up as decisiveness.

 

Retiring the Marshmallow Test

Any conversation about patience eventually reaches for the marshmallow test, the classic finding that children who delayed eating one marshmallow to get two later did better in life. A 2024 replication study in Child Development, using far larger and more representative data than the original, found the test doesn’t reliably predict adult outcomes at all: nearly every regression-adjusted relationship between childhood marshmallow performance and adult achievement, health, or behaviour came back statistically insignificant. The honest version of patience research doesn’t lean on a famous but shaky finding about children. It looks at what patience actually costs adults making real decisions.

 

The Bias That Makes Patience Feel Wrong

Behavioural economists have a name for the instinct patience has to fight: action bias, the tendency to act even when the evidence says waiting is the better choice, because acting feels like doing the job and waiting feels like doing nothing. A widely cited study of professional goalkeepers facing 286 penalty kicks found this bias in its purest form: goalkeepers dive left or right the overwhelming majority of the time, even though staying in the centre of the goal is statistically the better strategy, since kicks go there roughly 29% of the time. Goalkeepers dive anyway, because conceding a goal while standing still feels worse, and looks worse, than conceding one while visibly trying.

Medical research on “intervention bias” finds the identical mechanism in physicians: doctors reliably feel more satisfied recommending a treatment than recommending watchful waiting, “giving a sense of greater activism in their patients’ care,” even when the evidence supports doing nothing. Neither of these is really about competence. Both are about how much easier it feels to have visibly acted, regardless of whether acting was actually correct.

 

What Patient Leadership Actually Looks Like

A qualitative study of leaders who are known for patience found something specific underneath the trait: patience functions as a decision-making framework in its own right, not an absence of one, guiding a distinct process for weighing a situation before committing to act on it. A separate survey of 578 working professionals found leaders rated as more patient saw their teams’ self-reported creativity and collaboration rise by an average of 16%, and productivity by 13%.

A six-year study of more than 20 pairs of executives working in genuinely volatile markets coined the useful term for what this looks like in practice: active waiting. Not paralysis, and not indecision. Deliberate preparation during a lull, so that when a real opportunity actually opens, the leader can act decisively rather than reactively.

 

Why the Harder Choice Rarely Gets Made

None of the research above suggests patience is passivity’s better name. It suggests the opposite: patience requires resisting a bias that’s actively working against it in the moment, on a soccer pitch, in an exam room, or in a boardroom under pressure to be seen doing something. The impulsive decision is usually the one that requires less discipline, not more, because it removes the discomfort of visibly not acting while everyone is watching.

 

The Actual Choice Worth Making

The next time waiting is the objectively better move and doing something still feels necessary, the honest question isn’t whether patience is the right call. The research says it usually is. The question is whether you can tolerate looking like you’re doing nothing for exactly as long as doing nothing is correct.

Pre-Mortem: Twenty Million Members, No Published Error Rate

Posted on by

The Pre-Mortem is a weekly series on this blog. Each piece applies five questions to a major technology commitment before the outcome is known.

By the end of this year, twenty million Americans will use an AI companion to check whether their treatment has been approved, understand their benefits, and find out where they stand on a coverage dispute. UnitedHealth Group, the largest health insurer in the United States, calls it Avery. What the company has not published is what happens when Avery gets it wrong, and who carries it.

The Bet

UnitedHealth Group is investing more than $1.5 billion in AI in 2026. Avery is one part of a portfolio of over one thousand AI applications now operating across its insurance, pharmacy, and healthcare delivery businesses. The company expects a two-to-one return, much of it within the next eighteen months.

The scope goes further than the navigation functions Avery handles publicly. UnitedHealth has stated its intention to embed AI across claims decisions, clinical documentation, billing code selection, and fraud detection. The bet is that AI can absorb these regulated, high-stakes workflows faster than the accountability architecture around them can be clarified.

The Assumption

The whole bet turns on this, that an AI companion helping members find their benefits is categorically different from an AI algorithm making coverage decisions.

That distinction matters to UnitedHealth and to the regulatory debate around it. It is also the exact point where the accountability gap lives. Avery’s scope includes claim approval status and benefit explanations. In the sequence of a denied treatment, those interactions are not neutral, they are the moments where a member either understands their rights or does not. The line between navigation and decision sits precisely where the product is deployed.

 

The Sequence

UnitedHealth has been here before. Between 2019 and 2022, its subsidiary naviHealth deployed an AI tool called nH Predict to manage post-acute care decisions for Medicare Advantage members. A Senate investigation found that UnitedHealth’s denial rate for post-acute care claims more than doubled after nH Predict was deployed. A federal class action, Lokken v. UnitedHealth Group, alleges that the algorithm overrode treating physicians’ recommendations and carried a 90 per cent error rate on appeal, nine of every ten denied claims reversed when challenged.

That lawsuit is still advancing. In March 2026, a federal court ordered UnitedHealth to disclose its AI denial algorithm documentation, including internal AI Review Board materials, documents related to government investigations, and business records reaching back to 2017. Avery launched the same month to 6.5 million members, with a target of 20.5 million by year-end.

The sequence matters. The error rate history of the predecessor tool is documented and in litigation. The commitment not to repeat it with Avery has not been published in measurable form.

The Pager

UnitedHealth states that Avery is governed by a responsible use policy with review and approval from its AI Review Board. That board governs model development. No published framework names which specific individual, body, or governance layer is accountable when an Avery interaction contributes to a coverage outcome that causes patient harm.

The regulatory picture does not close that gap. At least twenty-five states have issued guidance under the National Association of Insurance Commissioners (NAIC) model bulletin. Alabama, Indiana, Washington, and others have enacted specific laws requiring human sign-off on AI-assisted denials, most taking effect in 2026. But the Employee Retirement Income Security Act (ERISA) preempts state action against self-insured employer plans, which cover the majority of employer-sponsored insurance. Federal oversight through the Centers for Medicare and Medicaid Services (CMS) and the Department of Health and Human Services (HHS) covers Medicare Advantage but carries no published standard for AI liability in individual claim decisions. The accountability is distributed. No name is on it.

The Proof

The $1.5 billion figure is confirmed. No committed outcome measure has been published for Avery’s error rate, its impact on denial rates, appeal success rates under AI-assisted decisions, or any patient safety incident reporting cadence.

Per CMS disclosures filed March 2026, the first year the agency required public reporting, UnitedHealth’s prior authorisation denial rate was 16.3 per cent in 2025, 4.8 percentage points above the industry average of 11.5 per cent. The company announced in May 2026 that it will eliminate prior authorisation for 30 per cent of services by year-end. Whether that changes the AI-in-the-loop accountability question for the remaining 70 per cent has not been addressed.

The Verdict

If the governance architecture catches up, if AI Review Board accountability is mapped to individual outcomes, if state AI denial laws close the ERISA gap, and if a committed outcome framework for Avery is published and audited, then this is exactly what responsible AI deployment in healthcare should look like, a major operator taking the accountability question seriously under public and regulatory scrutiny.

Without all three, twenty million people are interacting with an AI system whose error rate is undisclosed, whose predecessor carried a 90 per cent reversal rate on appeal, and where no named human is accountable for what it tells them about their care.

The bet is bold. The architecture to carry the loss has not been built yet.

Sometimes ‘Fail Fast’is Just an Excuse for Poor Planning

Posted on by

 

The Misuse of ‘Fail Fast’
The idea of ‘failing fast’ has become deeply ingrained in business culture, particularly in the world of startups and technology. The concept encourages teams to experiment quickly, learn from mistakes, and adapt with agility. While this approach has its merits, it’s increasingly being used as a shield for poor planning and lack of accountability.

Executives and business leaders often find themselves grappling with projects that are rushed to market under the guise of ‘failing fast,’ only to face costly setbacks and frustrated stakeholders. The problem isn’t with the concept itself, it’s with how and when it’s applied.

 

The Hidden Cost of Failing Fast
Failing fast, when genuinely used to encourage innovation and learning, can be valuable. However, when it becomes an excuse for poor decision-making or inadequate preparation, it leads to wasted resources, damaged credibility, and missed opportunities.

Instead of carefully assessing market needs or thoroughly testing new solutions, teams are encouraged to push products or services into the market with the justification that failure is part of the process. This mindset creates a dangerous cycle where failure becomes expected rather than avoided, leading to organisational complacency rather than growth.

 

The Real Problem
The problem lies in the difference between strategic experimentation and careless execution.

  • Strategic Experimentation involves calculated risk-taking, where failure is a potential but managed outcome.
  • Careless Execution is when the ‘fail fast’ mantra is used to justify a lack of preparation, unclear objectives, and poor risk management.

Leaders need to distinguish between the two. When ‘fail fast’ is used to bypass due diligence, proper planning, or thoughtful strategy, it becomes a crutch rather than a tool for growth.

 

How to Fix It
To leverage the value of failing fast without falling into the trap of poor planning, consider these strategies:

  1. Define Clear Objectives
    Before starting any project or initiative, establish clear goals and key performance indicators (KPIs). This ensures that the failure or success of an initiative can be measured accurately, rather than relying on vague outcomes.
  2. Create a Controlled Testing Environment
    Failing fast should happen within a controlled environment where the impact of failure is limited. Pilot programmes, A/B testing, and controlled rollouts allow teams to learn without risking major setbacks.
  3. Encourage Smart Failures
    Not all failures are equal. Encourage teams to take calculated risks but hold them accountable for thoughtful execution. Failure should lead to actionable insights, not be used as a fallback for poor planning.
  4. Assess Risk Before Moving Forward
    Failing fast does not mean ignoring risk. Conduct a thorough risk assessment before launching any new initiative. Anticipate potential challenges and establish contingency plans.
  5. Balance Speed with Quality
    Speed matters, but not at the expense of quality. Establish internal benchmarks to ensure that the need for quick feedback doesn’t lead to subpar products or services.

 

Moving from ‘Fail Fast’ to ‘Learn Smart’
Failing fast should not be a justification for sloppy execution or rushed decision-making. The goal should be to create an environment where teams are empowered to experiment, but within a framework that supports thoughtful strategy and measured risk-taking.

When teams understand the difference between smart failures and careless mistakes, they can pivot quickly without compromising long-term success. The key is to learn fast, not just fail fast.

 

Time to Rethink ‘Fail Fast’
Failing fast isn’t inherently wrong, but it’s not a strategy, it’s an outcome. Leaders who embrace strategic experimentation while maintaining strong planning and accountability will find themselves better positioned to drive sustained success.

It’s time to stop hiding behind the idea of failing fast and start leading with a defined purpose.

 

AI Risk Management: Unlocking Innovation Without Compromise

Posted on by

Artificial Intelligence (AI) is doing much more than just changing how we do business, it’s redefining it. But while AI opens doors to innovation and growth, it also comes with risks that can’t be ignored. Data bias, cybersecurity vulnerabilities, compliance gaps, these aren’t just technical issues; they are business-critical challenges.

How do you manage these risks without stifling innovation?

The answer lies in taking a deliberate, proactive approach to AI risk management. When done right, it’s not just about avoiding pitfalls, it’s also about creating opportunities, building trust, and future-proofing your business.

1. Assess Risks Before They Become Issues

AI’s complexity makes risk inevitable, but unpreparedness is a choice. Here’s where it starts:

  • Define Your Use Cases: Where and how will AI be applied? What’s at stake if it fails?
  • Spot Vulnerabilities Early: From biased data to weak cybersecurity protocols, address weak points head-on.
  • Plan for the Unexpected: Have contingency plans in place. AI systems are only as strong as the scenarios they’ve been trained for.

When you identify risks upfront, you’re not just protecting your business, you’re building a foundation for trust.

2. Monitor AI as if Your Business Depends on It (Because It Does)

AI systems evolve as they’re exposed to real-world data. That’s both their strength and their vulnerability. Without constant monitoring, you’re flying blind:

  • Detect anomalies before they escalate.
  • Ensure your AI complies with ethical, legal, and operational standards.
  • Create feedback loops for continuous improvement.

Think of it as a health check for your AI, one that keeps your systems resilient and your stakeholders confident.

3. Build a Workforce That Understands AI Risks

AI is powerful, but it’s only as ethical, secure, and effective as the people managing it. Here’s how you empower your teams:

  • Train them to recognise and mitigate risks at every stage.
  • Foster a culture where AI isn’t feared but embraced responsibly.
  • Equip employees with the tools to ask critical questions, like “Is this system fair?” and “What could go wrong?”

Knowledgeable teams are your first line of defence, and your greatest asset in turning risk into opportunity.

4. Stay Ahead of Regulatory Changes

AI governance is evolving faster than many realise. Falling behind isn’t an option. Stay agile by:

  • Keeping up with global and regional regulations.
  • Adapting processes to meet compliance requirements.
  • Engaging with industry groups to influence ethical AI standards.

Compliance isn’t just about ticking boxes; it’s about positioning yourself as a trusted, forward-thinking leader in AI adoption.

5. Embed Trust with AI TRiSM

The AI Trust, Risk, and Security Management (TRiSM) framework is a revolutionary approach to ensuring your AI systems operate securely, ethically, and effectively. It achieves this by:

  • Protecting data integrity and maintaining model accuracy.
  • Shielding systems from adversarial attacks.
  • Ensuring your AI aligns with your ethical and operational values.

By embedding TRiSM principles, you not only safeguard your operations but also build a foundation of trust that resonates with stakeholders and sets your organisation apart as a leader in responsible AI innovation.

6. Make Risk Awareness Part of Your Culture

AI risk management isn’t a one-off task, it’s a mindset. Leaders must lead by example:

  • Make risk conversations part of regular strategy discussions.
  • Encourage collaboration between technical and non-technical teams.
  • Celebrate transparency and accountability, acknowledging risks isn’t a failure; ignoring them is.

A culture that prioritises awareness over avoidance turns AI risks into stepping stones for growth.

Turning Risks Into Rewards
Let’s shift the narrative: AI risk management isn’t about fear. It’s about foresight. When you manage risks effectively:

  • Your business earns trust, from customers, stakeholders, and regulators.
  • ou realise the full potential of AI, without compromise.
  • You gain a competitive edge by showing you can innovate responsibly.

It’s time to stop seeing risk management as a hurdle and start seeing it as a strategic advantage.

The world is moving fast, and AI is at the centre of it. But you can’t afford to sit back and hope for the best.

So, ask yourself:

  • Are your AI systems being monitored for vulnerabilities right now?
  • Are your teams trained to manage AI risks effectively?
  • Do you have a plan for when, not if things go wrong?

Managing AI risks isn’t just about protecting what you’ve built, it’s about creating what comes next. The organisations that get this right will thrive.

 

The working hour week

Posted on by

Some years ago in my old blog I posed this question. What is the ideal working week around the world.

I was discussing the working hour week with friends that work around the world in different countries, comparing the average and what is now deemed as acceptable or viewed as the new standard, regardless of the contracted hours  so these posts caught my eye.

Some people feel they have to work long hours or stay late at work even when there is nothing to do. Almost like staying late is the norm, to the extent of faking it. Then there is the other side that seem to think there should be some shame associated with working long hours. 
An old survey lists out which countries work the longest hours in Europe the real question though is how does this link to productivity.

Personally I think it’s all about having a balance and scaling up and down as required rather than having a set pattern. However it looks like, due to Covid forcing organizations to make workers remote all this is about to change. There have been a few news reports about burnout and even longer working hours but this time whilst working from home.

Even the WHO has jumped in raising the awareness of increasing deaths from heart disease and stroke. “Working 55 hours or more per week is a serious health hazard,” said Maria Neira, the director of the WHO’s Department of Environment, Climate Change and Health.

There is also talk about reducing the working week to 4 days as seen on CNBC and the 4 Day Week Campaign here

What do you think?