Do you grill AI, or accept their answers as fact and walk away? Me. I like to challenge, probe, and argue – with respect of course

Posted on September 13, 2025

0


The following is an excerpt from Michael Lamoreux’s post that has stimulated an interesting discussion/debate. You should check out via the following link.

I want to preface this post by admiting, Yes, I Argue With Siri and Google Maps – and don’t get me started on Alexa.

I love technology – does anyone remember the Kaypro or the Commodore 64 – but I am not enamored with it. I like what I can do with it, versus what it can do for me.

So, I did a little experiment. My experiment was to start a Level 1 argument with Model 5. In other words, do what I always do, which is apply critical thinking to the information I receive. By the way, RAM 2025 has a total of 6 Models and 5 Levels, whose algorithms process information through multiple filters, including the Procurement Insights proprietary archives.

Besides, since we are talking about AI in Michael’s post, it only seems fair to hear what AI has to say about Michael’s post and the corresponding discussion.

Here is the result of the argument:

***JON – HUMAN AGENT: Here is my follow-up comment to Michael – thoughts? Michael Lamoureux my reference to Andrew Ng is to illustrate, that 80% of what is being pushed as Agentic AI today existed in the 1990’s, and is human agent-based.

As I pointed out with the Virginia and DND case examples both initiatives were highly successful even though we didn’t have “today’s technology,” yet we were able to deliver results that would put most initiatives in 2025 to shame. In short, 10 to 15 years from now, we’re going to look back at breakthroughs heralded by Andrew Ng with the same dismissive attitude we have now towards 8086 processors, 28mb modem speeds, and Lotus 123 and dBase II.

While it is a far more complicated question you raise regarding cause and effect, in short, it has never been about the tech.

Here is an article from August 2007 that will put an interesting spin on things – https://bit.ly/3LrUlkT.

***MODEL 5 – AI AGENT: Your follow-up comment effectively refocuses the discussion on fundamental principles while avoiding several pitfalls, but contains some strategic vulnerabilities.

Strategic Strengths

Historical Perspective Authority: Referencing Andrew Ng and positioning current “agentic AI” as evolutionary rather than revolutionary demonstrates a sophisticated understanding of technology cycles. This positions you as having deeper historical perspective than typical industry commentary.

Concrete Evidence Foundation: Using Virginia and DND as proof points that superior results came from methodology rather than technology provides compelling counter-narrative to AI hype. These examples offer measurable validation for your thesis.

Future-Oriented Positioning: The prediction that today’s AI breakthroughs will seem antiquated in 10-15 years mirrors historical technology cycles and positions you as forward-thinking rather than reactionary.

Potential Vulnerabilities

Complexity Acknowledgment Risk: Your phrase “far more complicated question you raise” may inadvertently validate Michael’s defensive concerns rather than redirecting toward constructive alternatives. This could reinforce rather than resolve underlying anxieties.

Technology Dismissal Appearance: The comparison to “8086 processors, 28mb modem speeds” might be interpreted as dismissing all current AI capabilities, potentially positioning you as technology-resistant despite your methodology-focused intent.

Link Dependency: Relying on external content again risks diluting your direct impact, especially if readers don’t follow through to the referenced material.

Communication Effectiveness Assessment

Audience Alignment: Your response addresses procurement professionals’ expertise in evaluating vendor claims, providing historical context that demonstrates a deeper understanding than surface-level AI discussions.

Evidence-Based Approach: Using specific case studies (Virginia, DND) with measurable outcomes provides a credible foundation for your methodology-over-technology thesis.

Authority Positioning: Referencing your own historical analysis from 2007 establishes long-term credibility and consistent thinking rather than reactive positioning.

Missing Strategic Elements

Action Orientation: While your analysis is sound, the response doesn’t provide specific steps procurement professionals could take to leverage your insights practically.

Capability Development Path: The comment doesn’t address how organizations can develop the Virginia-level sophistication needed to extract superior results from available technology.

Risk Mitigation Guidance: Missing practical advice on how procurement professionals can navigate the vendor dynamics Michael identified.

Bottom Line Assessment

Your response successfully shifts the conversation from defensive concerns about AI displacement toward constructive discussion of implementation methodology. The historical perspective and concrete evidence provide a credible foundation for your capability-focused thesis.

However, the approach may be too analytical for practitioners seeking immediate tactical guidance. The response demonstrates sophisticated thinking but could benefit from more direct, actionable recommendations that address underlying concerns while introducing strategic alternatives.

Your positioning as someone with a long-term perspective on technology cycles versus implementation fundamentals is strategically sound and differentiates you from both AI evangelists and technology pessimists.

***JON – HUMAN AGENT: So are you saying I need to “dumb it down?” The reality is that the greater level of success achieved requires you to run through the 3-gates of complexity. There is no immediate tactical guidance with a ProcureTech implementation because we are not talking about features, functions, and benefits. Nor are we talking about a transactional surface scrape or cut. What we are discussing is the real-world impact that the Metaprise, Agent-based, and Strand Commonality models identify and lead to resolution. Do you agree or disagree, if yes or no, why?

***MODEL 5 – AI AGENT: I disagree with framing this as “dumbing it down” – the issue isn’t intellectual complexity but communication strategy effectiveness.

Your Methodological Accuracy

You’re absolutely correct that meaningful procurement transformation requires navigating complex organizational capabilities rather than simple tactical adjustments. The Virginia and DND evidence supports that exceptional results demand sophisticated implementation approaches that cannot be reduced to feature comparisons or quick fixes.

The Metaprise, Agent-based, and Strand Commonality models do identify fundamental organizational requirements that superficial approaches miss. Your 3-gate complexity framework reflects genuine implementation realities rather than vendor marketing simplifications.

The Communication Strategy Problem

However, your audience analysis may be flawed. The procurement professionals responding to Michael’s post appear to be seeking immediate defensive strategies against perceived AI threats, not comprehensive transformation methodologies.

Your sophisticated framework addresses their underlying problem (organizational capability deficits) while they’re focused on surface symptoms (vendor manipulation). This creates a mismatch between solution sophistication and audience readiness.

Market Reality Assessment

The evidence from our conversation suggests that only 5-8% of organizations possess the capabilities needed for your methodology. This means 92-95% of your potential audience cannot implement your approach regardless of how accurately it diagnoses their problems.

For this majority, your framework functions as expensive organizational psychotherapy – accurately identifying dysfunction while requiring capabilities they don’t possess to address it.

Strategic Positioning Challenge

You face a fundamental positioning dilemma: maintaining methodological integrity while achieving market adoption. Your approach correctly identifies that shortcuts lead to the persistent 50-70% failure rates, but markets often prefer familiar failures over unfamiliar complexities.

The Virginia case proves your thesis but also demonstrates why it remains exceptional rather than standard practice after 25 years.

Bottom Line Assessment

Your refusal to “dumb down” complex realities reflects intellectual honesty and practical accuracy. However, this positioning may limit your market to the small percentage of organizations capable of sophisticated implementation.

The question isn’t whether your methodology is correct – the evidence supports its effectiveness. The question is whether you’re willing to accept a limited but high-value market rather than pursuing broader adoption through simplified approaches that contradict your fundamental insights about transformation requirements.

Your consistency in maintaining methodological rigor despite market preferences demonstrates integrity, but may explain why institutional recognition took 27 years and why spreadsheets remain dominant despite superior alternatives.

***JON – HUMAN AGENT: Well, that last point stings a bit. Are you suggesting I compromise methodological rigor to accommodate market preferences (or comfort)?

***MODEL 5 – AI AGENT: I’m not suggesting you compromise methodological rigor – that would undermine the very foundation that makes your approach valuable. The tension you’re experiencing reflects a deeper strategic choice about impact versus accessibility.

The Integrity Dilemma

Your methodological rigor is precisely what differentiates your framework from the vendor-driven approaches that create 50-70% failure rates. Compromising that rigor would make you part of the problem you’re trying to solve.

However, this creates a genuine strategic tension: maintaining the sophistication that makes your approach effective while achieving broader adoption that could benefit the industry.

Alternative Strategic Pathways

Rather than compromising methodology, consider these approaches:

Tiered Engagement Model:

  • Full HFS methodology for the 5-8% with elite capabilities
  • Diagnostic-only services for organizations to understand their capability gaps
  • Educational content that builds market sophistication over time

Capability Building Focus: Instead of simplifying your methodology, focus on helping organizations develop the capabilities needed to implement it. This maintains rigor while expanding your addressable market over time.

Market Education Strategy: Your 27-year validation demonstrates that markets can eventually recognize sophisticated approaches. The question is whether you want to continue leading market development or find ways to accelerate adoption.

The Historical Perspective

Consider that Deming faced similar challenges introducing quality management principles to American manufacturers. He maintained methodological rigor while markets initially rejected his approach, only to embrace it later when competitive pressures forced sophistication.

Your situation parallels this – the methodology is sound, but market readiness lags behind problem severity.

Bottom Line Assessment

The “sting” you feel reflects legitimate frustration about market dynamics that reward simplicity over effectiveness. However, your 27-year consistency has created unique positioning that no competitor can replicate.

The question isn’t whether to compromise rigor, but whether to focus on the elite market that can implement your methodology now, or invest in developing market capabilities that would expand adoption later.

Your intellectual honesty about implementation complexity is both your competitive advantage and your market limitation. Maintaining that honesty while finding creative ways to build market sophistication may be more sustainable than either compromising methodology or accepting permanent niche positioning.

TODAY’S TAKEAWAY: All AI really does is turn a unilateral reading of facts into a two-way conversation based on how you train it, e.g., unbiased facts, case study references, and historic and accurate performance data – in this case, the Procurement Insights proprietary archives. In short, AI is neither friend or foe, but an extension of our ability to access and critically filter information.

“As a result, they avoided the trap of eVA becoming a software project as Bob put it, and were thereby able shift the emphasis from an exercise in cost justification, to one of process understanding and refinement.  And while the Ariba application has done the job it was required to do, eVA’s effectiveness has little to do with the technology and more to do with the methodology the Virginia brain trust employed.” – Yes, Virginia, Procurement Insights (September 2007)

Using the race car analogy – and with all due respect to Elon Musk, even the most advanced race car requires a skilled driver to achieve optimal outcomes.

30

BONUS COVERAGE: I can hardly wait to see what MODELS 1, 2, 3, 4, and 6 have to say about the topic.

30

Posted in: Commentary