Joël Van Hoolandt — Fractional CTO | Magma Consulting

Some companies dream of a quiet, orderly march toward their IPO. Anthropic, it seems, prefers the scenic route. The firm filed confidential IPO paperwork with the SEC on 1 June 2026, reportedly putting it on track to be the first AI company to test public markets at a valuation flirting with the trillion-dollar mark. A fine moment to project calm competence — which is, of course, precisely when the universe tends to schedule its plot twists. On 9 June, Anthropic launched Fable 5 and Mythos 5; three days later, both models were offline everywhere in the world.  The cause was not a glamorous corporate scandal but something more bureaucratically dramatic: the U.S. Commerce Department invoked national security export controls to bar the company from distributing the models to any foreign national — a category that, awkwardly, included Anthropic's own non-citizen employees. The directive landed at 5:21 p.m. Eastern on a Friday, because regulators apparently share my fondness for ruining...

  In our line of work, standing still is the quickest way to fall behind. The technology landscape reinvents itself roughly every eighteen months, and clients quite reasonably expect their consultants to have read the memo — ideally before they did. That is why we have invested, once again, in a fresh round of training and certifications across cloud, AI, and modern infrastructure. Think of it as keeping our toolbox sharp, our methods current, and our coffee consumption heroically high. Certifications, of course, are not trophies to be admired on a shelf (though we will admit they look rather nice on LinkedIn). Their real value lies in what they enable: faster diagnostics, sounder architecture decisions, and the quiet confidence that comes from knowing the difference between a passing trend and a foundation worth building on. Each badge represents hours of study, a few late evenings, and the occasional muttered conversation with a documentation page that refused to cooperate. Wha...

Across Europe, public administrations are being asked to do more with less: ageing populations, rising service expectations, and tight fiscal constraints have become the norm rather than the exception. Digital transformation, once about putting forms online, now blends automation with AI to unclog the back office and simplify the citizen’s journey. Case files can be summarised in seconds, triaged more fairly, and translated across languages without losing nuance; routine verifications can be automated with auditable trails, not opaque shortcuts. The productivity story is not only about speed, but about quality, consistency, and the space for human judgment where it matters most. For the workforce, this is not a tale of mass replacement but of work changing shape. Expect the routine to thin out and the complex to thicken, with public servants spending more time on interpretation, outreach, and decisions that require empathy and proportionality. New roles are already appearing—product...

When European boards talk about “AI sovereignty,” they are usually wrestling with a quieter, more practical anxiety: not where the tin sits, but who can compel access, who can read the logs, and how easy it will be to unwind a choice when the world shifts. Compliance is a threshold, not a destination ; it tells you whether you may proceed, not whether you are in control. The sharper conversation is about jurisdiction layered over routing, about telemetry as much as inference, and about exit paths that don’t break the product. In that light, choosing between Claude on Bedrock and GPT on Azure is less about model poetry and more about plumbing you would be comfortable defending to a regulator and explaining to a customer. On AWS, Anthropic’s Claude through Bedrock is a good example of EU residency done with a clear spine: deploy via Ireland or Stockholm and requests stay within a single region; pick Frankfurt or Paris and you’ll use the EU cross‑region inference profile, keeping data w...