Best of LinkedIn: Artificial Intelligence CW 21/ 22

There is a clear shift in the market conversation: AI is no longer discussed primarily as experimentation, productivity, or tool adoption. The dominant themes are governance, agent readiness, operating model change, cost control, and trust. For account-based marketing, the implication is clear: enterprise buyers now need sharper proof of control, compliance, business impact, and implementation credibility.

Date

June 4, 2026

Artificial Intelligence

Thomas Allgeyer

Methodology: Every two weeks we collect most relevant posts on LinkedIn for selected topics and create an overall summary only based on these posts. If you´re interested in the single posts behind, you can find them here: https://linktr.ee/thomasallgeyer. Have a great read!

‍

Listen to our podcast

If you prefer listening, check out our podcast summarizing the most relevant insights from Artificial Intelligence CW 21/ 22:

Regulatory Pressure and AI Governance

EU AI Act became the dominant market signal, framed as a global benchmark rather than a Europe-only compliance topic
Scope, transparency, Article 50, HR-Tech exposure, and August 2026 readiness emerged as recurring concerns for boards and operators
Runtime governance positioned as the next maturity layer, moving beyond static policies, review boards, and documentation
Compliance evidence increasingly expected at the decision level, including authorization, policy checks, audit trails, and execution context
Human-in-the-loop framed as insufficient when AI systems materially influence employment, finance, insurance, or other regulated decisions
Global governance landscape shown as fragmented, with signals from the EU, US litigation and SEC enforcement, Singapore, China, UAE, Pakistan, and copyright regimes

‍

Agents and Architecture

AI agents highlighted as the next enterprise adoption wave, but with trust, permissioning, escalation, memory, and control as core blockers
Microsoft Build, agentic AI governance frameworks, and enterprise adoption guides reinforced that outcome design matters more than tool selection
Agent memory emerged as a critical architecture topic, covering working memory, episodic memory, semantic memory, and procedural memory
Chain-of-thought monitorability positioned as a key engineering challenge for trustworthy agentic systems
Bol’s Billie agent showed a concrete commerce shift, where product discovery moves from search results to single-SKU recommendation
Agent governance increasingly tied to what systems can do in deployment, not how they are technically built

‍

Enterprise Adoption and Operating Model

Enterprise AI discussion shifted from “which tool” to “which workflows, roles, decisions, and controls need redesign”
Data quality and availability identified as a recurring bottleneck that prevents AI from scaling beyond experimentation
AI exposed organizational maturity gaps, including weak process ownership, unclear accountability, and insufficient operating discipline
Leaders framed AI implementation as an operating system question, requiring direction setting rather than simple task allocation
Workforce redesign became more visible, with entry-level roles, institutional knowledge, and capability building identified as strategic risk areas
AI literacy and training gaps emerged as major blockers, especially where employees use AI without structured enablement

‍

Cost, Compute and Infrastructure

Compute scarcity positioned as a strategic constraint, with Anthropic and xAI illustrating how infrastructure access shapes competitive advantage
GitHub’s token-based Copilot billing highlighted the end of subsidized AI usage and the need for stronger cost architecture
Unity AI Gateway introduced spend controls as a practical response to uncontrolled token consumption across enterprise AI tools
On-device AI gained attention as a serious architecture option, especially where privacy, latency, unit economics, and user experience matter
Sovereign AI discussion moved from location-based hosting to hardware-backed security and application-layer encryption
AI infrastructure increasingly framed as critical national and corporate infrastructure, not just a technology stack

‍

Trust, Ethics and Human Judgment

AI ethics discussed as a business operating principle, not only a regulatory checkbox
Human dignity, institutional dependence, and responsible use appeared across posts linked to religion, public policy, healthcare, and development justice
Several posts stressed that human connection, skepticism, and judgment remain differentiators in AI-driven business environments
AI literacy positioned as both a corporate and civic priority, including examples from Malta, webinars, and graduation-stage commentary
Layoffs, automation risk, and entry-level role disruption surfaced as early signs of broader workforce tension
The strongest governance argument was not fear-based compliance, but trust as a prerequisite for adoption and market access

‍

Products, Partnerships and Market Moves

Bol introduced Billie as an AI shopping agent that can select and add products directly to cart
Unity AI Gateway launched AI spend controls to give enterprises visibility and budget governance across AI usage
GitHub shifted Copilot toward token-based billing, raising enterprise attention on AI cost transparency
Malta partnered with OpenAI to provide free ChatGPT Plus access, combined with mandatory AI literacy training
Deloitte was recognized again as an IDC MarketScape Leader in AI Services, reinforcing demand for enterprise-scale implementation support
Singapore’s Agentic AI Governance Framework stood out as one of the more practical enterprise-ready governance references
Mendo positioned enterprise GenAI adoption as a C-suite discussion, bringing CIOs, CHROs, Microsoft, Verallia, and Edenred into one forum

‍

Subscribe to newsletter

Subscribe to receive the latest blog posts to your inbox every week.