OP-ED - What good navigation looks like: no machine sails itself
From Block's standing ovation to Monday.com's manifesto to a South African consultancy that started with the social system, Part 3 asks what good AI navigation actually looks like when you prioritise the human element and demand results over rhetoric.
Over the past two weeks, this column has argued two things.
Part 1 made the case that the AI conversation has a prediction problem: we are so busy forecasting the future that we have lost the habit of reading our actual horizon.
Andile Masuku
Part 2 made the case that we have been here before: every major technology deployment that ignored the social system alongside the technical one produced the opposite of its intended outcome. The coal mines told us that in 1951. IBM's jarring 13.2% share-price drop in a single day a couple of weeks ago confirmed the machines are arriving again.
Andile Masuku
Part 3 is the harder question. If good navigation is not prediction, and if technology without social redesign is a fairly predictable path to waste, what does a well-calibrated response actually look like? The bar I apply is simple: demonstrate a customer outcome, or do not expect my serious attention. Frameworks and manifestos are a dime a dozen. I am after results.
The cut that got a standing ovation
By that standard, Jack Dorsey's decision to cut Block nearly in half (from over 10,000 employees to just under 6,000) deserves scrutiny before it deserves applause.
The market certainly applauded. Block's share price jumped 15%. Dorsey's framing was careful: "We're not in trouble. But something has changed." The severance terms were generous by corporate standards. You can read that as an organisation that understood its social obligations precisely because it understood what it was asking people to absorb.
There is a less flattering read. Block tripled its headcount during and after the pandemic, growing from roughly 3,800 employees in 2019 to over 12,000 by 2022. It was slower than its peers to scale back. Its stock had fallen roughly 40% since the beginning of 2025. It’s a trajectory that had nothing to do with AI and everything to do with a business that had grown unwieldy. Bloomberg has already posed the awkward question: is this AI transformation, or is it AI-washing? That is, cost-cutting dressed as technological futurism.
Dorsey, to his credit, addressed the criticism on X. He acknowledged building two separate company structures for Square and Cash App, a mistake corrected only in mid-2024. Block is now targeting over USD 2 million in gross profit per employee, quadruple its pre-pandemic figure. That response is more honest than the original letter.
yes we over-hired during covid because i incorrectly built 2 separate company structures (square & cash app) rather than 1, which we corrected mid 2024. but this misses all the complexity we took on through lending, banking, and BNPL. and that we’re now targeting $2M+ gross… https://t.co/uaKgTdzGob
— jack (@jack) February 27, 2026
It is also, inadvertently, the strongest evidence for the AI-washing read. A company that only corrected a structural mistake in 2024 is not describing AI transformation. It is describing a belated restructuring that coincides with the AI narrative.
The pattern extends beyond Block. A Forbes analysis mapped the broader dynamic: UPS cuts 48,000 jobs, stock rises. FedEx announces the closure of over 475 facilities, stock rises. The intent to automate has become a valuation catalyst before the automation itself is functional. Most of the real economy (in Africa and much of the rest of the world) is not a SaaS company. It does not move at venture capital speed. And the hyper-extractive logic that rewards share price bumps disconnected from actual business fundamentals is not a compass anyone should be navigating by.
The manifesto and the product
Notion announced "Custom Agents" in late February. The marketing copy reached for significance: "The AI era should leave no one behind." That is a manifesto, and I am tired of manifestos. They are not evidence. What the product actually offers is more interesting than the framing: autonomous agents that run without code, set up in minutes, operating across business applications, switchable between models. One person builds it; the whole organisation benefits.
At @NotionHQ, we believe every business deserves powerful & beautiful tools.
— Ivan Zhao (@ivanhzhao) February 24, 2026
AI is the most important technology of this era. It shouldn't only belong to companies that have AI teams or can afford forward deployed engineers.
Today we're launching Custom Agents:
— The first… pic.twitter.com/0PS77W7XdA
For African enterprises specifically — businesses that cannot afford forward-deployed engineers, that operate across mixed-connectivity environments, that have been priced out of the consultant-and-implementation model for decades — this matters if and only if it works as described. The promise of accessible automation has a long and inglorious history of not surviving contact with real operational conditions.
There is also the small matter of the walled garden vibes: Notion's value proposition asks you to adopt its platform architecture wholesale, or not at all. For organisations already running fragmented systems, that is not a minor caveat. The direction is right. But direction is not arrival, and arrival is not a customer outcome.
The CEO who said the quiet part out loud
Monday.com CEO Eran Zinman's recent interview with venture podcaster Harry Stebbings struck a different chord. Monday.com (a work management platform with aspirations not unlike Notion, with over USD 1.3 billion in annual recurring revenue) is valued at roughly USD 3.9 billion. That’s more than 60% decline since its IPO in 2021. Zinman described watching the stock fall to USD 70 and arriving at a peculiar clarity: "What the market is telling me is the company's worth zero. Fine. Now I need to build."
Under pressure from Stebbings — who appeared rather gleeful pushing the case for why Monday.com had not already sacked half the company — Zinman did not (entirely) retreat into platitudes. And some of what the company has done passes the bar.
Monday.com has replaced its entire sales development representative (SDR) function with AI agents. Response times went from 24 hours to three minutes and conversion rates went up. The displaced SDRs were moved into account management. Customer support is now AI-driven. These are real operational changes producing measurable results.
But the broader pitch drifted into territory I am tired of. Zinman's vision for Monday.com as the default platform where humans and agents collaborate across the organisation is a manifesto. An ambitious one. Possibly even a correct one. But it is not a demonstrated outcome. Zinman came across as more hopeful than pragmatic. And at this point in the cycle, hope is not a horizon scan. It is a forecast. And we have spent two columns establishing why forecasts can be the wrong instrument.
The outfit that started with the social system
Which is why the most compelling navigation story in this series comes from closer to home. Superworker is a platform built around a straightforward proposition: the most important question in the intelligence era is not how many people you can replace, but how quickly you can grow the capability of the people you have.
It was spawned by VSLS, a South African consultancy with more than two decades of change management experience and a growing international footprint. This is not a startup that arrived at workforce augmentation as a trend. It arrived there through 20 years of watching what happens when organisations acquire capability without redesigning the conditions under which people use it. The coal mine lesson was built into the founding instinct (and, admittedly, in response to an existential threat to their own legacy consulting model).
PKF Australia's recent deployment of Superworker will be worth examining over time for real outcomes. A professional services firm no doubt facing the same disruption that IBM's Monday laid bare — and apparently choosing to grow its people's capability and productivity rather than simply defaulting to reducing headcount. The social system and the technical system, designed together. Not a manifesto. A working deployment.
What is the role of technology as 500 million African youth reach working age over the next 30 years? – Jake Kendall (Editor, DFS Lab)
The alarm, not the forecast
The 500 million jobs question that investor and researcher Jake Kendall asks about Africa's young workforce does not resolve into optimism or pessimism. It resolves into navigation.
The human element in an organisation is not an inefficiency to be engineered away. It is infrastructure. Remove it before you have redesigned the system around it, and you do not get a leaner company. You get a company that has forgotten how it works.
The sailors who survive (and thrive) are not the ones who predicted the storm. They are the ones who understood that no machine sails itself.
Editorial Note: An abridged version of this opinion editorial was first published by Business Report on 10 March 2026.