Analytical briefings for engineers, PMs, and CTOs
7 min readEnterprise AI vendors are replacing per-seat licences with consumption-based pricing — transferring compute cost volatility to customers. Gartner finds organisations can make 500–1,000% errors in AI cost calculations without proper modelling.
6 min readDocument AI and audio transcription are in production at scale. Video understanding and open-ended visual reasoning are still in pilot. A modality-by-modality breakdown of where multimodal AI has earned its place in enterprise workflows — and where reliability gaps are keeping CFOs from removing human review.
7 min readA documented sequence of confirmed attacks across 2025 — including zero-click data exfiltration via Microsoft 365 Copilot and malicious MCP servers distributed through npm — defines a new category of enterprise exposure. Here is what the incident record shows and what security teams must do now.
3 min readGPT-o3, Gemini 2.5 Pro, and DeepSeek R1 are in production — but at $15–60 per million output tokens, costs are 8–40x higher than standard models. Most organisations have not rebuilt their infrastructure assumptions to account for it.
4 min readControlled studies show AI coding assistants reduce time-to-completion by 26–55% — but gains concentrate in code generation and do not extend uniformly to debugging, architecture, or code review. The contexts matter enormously.
3 min readMicrosoft, Google, Amazon, and Meta will collectively spend over $320 billion on AI infrastructure in 2026 — but more than 60% is going into power infrastructure and data centre construction, not compute hardware. The binding constraint has shifted from chips to electricity.
4 min readEnterprise AI ROI is real but concentrated: financial services, healthcare administration, and software development show payback periods under 18 months. Forty-one percent of enterprise deployments remain in pilot phase after more than 12 months.
4 min readThe US has no federal AI law, 40+ competing state bills, and executive orders that reversed within two years. For enterprises deploying AI in hiring, credit, and healthcare, legal exposure exists now under existing anti-discrimination and consumer protection law — not future AI-specific regulation.
4 min readThe AI industry has no consensus definition of ‘production’ for AI agents. Most enterprise deployments labelled ‘production’ are narrow automations with human oversight, not autonomous multi-step systems. The definitional gap creates real risk: procurement decisions, SLA commitments, and compliance claims are built on undefined terms.