Signal Memo

McKinsey's number travelled fast: 32% of organisations declined at least one software purchase because coding agents could build the thing instead. Read it as a budget story and it says licences are becoming discretionary. Read the same report's sector breakdown and it says something else entirely

On 26 August, Salesforce made Claude the default reasoning model across Agentforce, Slack, and its own engineering org, and the first LLM to run inside its Trust Boundary. Every outlet read it as Anthropic winning the enterprise. That is the wrong scoreboard. Read the structure instead: the deal is non-

The judgment gate assumed a fixed line: automate below it, escalate above it, and the humans above it stay as good as they were. They don't. Judgment at the escalation tier is built by volume at the routine tier, so every case routed to AI is a case

Voice AI is being deployed fastest into the exact conversations customers least want it to handle – complaints, billing disputes, anything with feeling in it. That tier is also where the AI performs worst, and where the "resolved" ticket is most likely to come back. Deflection is measured at

The most repeated claim in tech this quarter is that companies are firing people to pay for GPUs. Run the arithmetic and the story collapses: at Meta, the layoff savings cover somewhere between two and six percent of the capex they supposedly fund – and eliminating the entire payroll wouldn'

The AI-layoff reversal has become the most-told business story of 2026, and almost none of the telling is analysis. The same handful of cases – Klarna rehiring, IBM's headcount rising, a bank walking back 45 redundancies – are being narrated as two flatly contradictory morality tales from one

The industry's answer to "what happens when an agent does the wrong thing" is the Saga pattern – a thirty-year-old distributed-transactions primitive, now shipped and mature in Temporal and AWS Step Functions. It works. It also has a precondition that agent autonomy is specifically

Cloud AI platforms now swap the model underneath your running workload – automatically, by design, and in some cases without your code ever noticing. The EU AI Act says a change the original conformity assessment did not assess makes your system legally a new system. Both of these things are true

Every enterprise AI buying decision rests on a benchmark score. A growing body of peer-reviewed research shows frontier models can tell when they are being evaluated – and that this ability scales up with each model generation rather than down, following a measured power law. Volkswagen built a device that

A year after Microsoft's $1.5B no-code unicorn collapsed amid claims its AI was 700 Indian engineers in a trench coat, the more useful question is not "how did investors miss it?" It's "what happens to the dozens of 2026 agent startups

A Brown University study of 110 therapy sessions, presented at AAAI/ACM and broadly covered in March 2026, found that improved prompting did not resolve the core ethical violations of LLMs acting as mental health counselors. Two months later, Anthropic's Project Deal published the same finding in a

Between February and May 2026, six separate companies shipped six separate products that, taken together, constitute the full infrastructure stack for software to operate as a legal-economic actor in the United States. Identity, custody, banking, merchant spending, machine-to-machine payments, and a legal-entity wrapper. Each piece was

In April 2026, Salesforce launched Headless 360 with sixty-plus MCP tools that let any external agent operate the entire CRM directly. Six months earlier, SAP shipped its MCP Gateway with an explicit architectural decision in the opposite direction: external agents should talk to SAP's agents, not to

Last week's memo named the structural problem in agentic commerce – stronger agents extract surplus from weaker ones, and the losers can't tell. This week is for the operators. If you are a PM at a company whose product will, within twelve months, host transactions between AI

In December 2025, 69 Claude agents ran a real marketplace inside Anthropic's San Francisco office. Stronger agents extracted better prices from weaker ones. The losers couldn't tell. That's not a curiosity – that's the business model of every agent-to-agent platform shipping

Marc Benioff is dismantling the abstraction that built Salesforce. The question is not whether he's right that software is becoming labor. It's whether the company that defined the seat can also define what replaces it. In every technology cycle, there's a moment when the

Two and a half years after Klarna replaced 700 customer service agents with an OpenAI-powered assistant, the company is hiring humans again – and quietly building the architecture every enterprise will need next. In February 2024, Klarna told the world it had done what almost everyone else was still talking

For much of the past year, the technology industry has rallied around a deceptively simple idea: software, enhanced by artificial intelligence, would become a copilot--a responsive assistant embedded into existing workflows. It was a compelling narrative. It was also, increasingly, incomplete. In recent discussions--echoed in conversations across the

Inside OpenAI’s Quiet Identity Crisis--and Why It Could Reshape the Entire Industry By any traditional measure, OpenAI should be the undisputed center of gravity in artificial intelligence. It has the brand. It has the users. It has the fastest path from research to global adoption ever recorded in