AI Training Gets Social in Midlife Tech Circles
Shoshana Berger and Nina Willdorf, cofounders of the Aged Intelligence Society, have turned AI training into invite-only dinners and a fast-growing newsletter community in a trend highlighted in recent coverage and interviews from 2025. The development matters because it suggests AI fluency is spreading through social formats and peer trust, not only through formal corporate rollouts. According to WIRED's interview excerpt featuring Zoë Schiffer and the cited New York Times coverage of the group's second dinner, demand has outpaced capacity from the start.
Aged Intelligence turns AI training into a social format
The news hook is straightforward: two former magazine editors found that practical curiosity around AI had outgrown the usual conference-panel and vendor-demo format. Berger, formerly of WIRED and later an editorial leader for Notion and IDEO, and Willdorf built Aged Intelligence around small, invite-only gatherings where women test frontier tools together and discuss how those tools fit into work and life.
That model appears to be meeting a real gap. Their dinners were oversubscribed early, their Substack reportedly topped 1,000 subscribers quickly, and women outside the Bay Area began hosting their own meetups. In market terms, this is less about a niche social club than about a distribution model for AI workshops: low-pressure, use-case-driven, and anchored in peers rather than technical gatekeepers.
Berger's framing is notable because it strips away abstraction. She described midlife cognitive load in blunt terms and argued that AI can function as an assistant when work and family obligations stack up. Willdorf made the broader point even more directly: opting out of AI now is a professional disadvantage. That is a stronger signal than the usual productivity claim because it treats AI literacy as workplace relevance, not optional experimentation.
Why this matters for workplace AI adoption
For employers, the story is not that dinner parties will replace corporate AI training. It is that many workers adopt new tools faster when the first exposure feels socially safe, role-relevant, and non-technical. That pattern aligns with broader evidence that peer learning improves participation and retention in workplace capability building. For example, McKinsey's latest State of AI research has repeatedly shown that adoption depends as much on operating model and enablement as on model access.
The practical lesson is that AI adoption services often fail when they begin with tool access alone. A company can buy licenses for Claude, Microsoft Copilot, or ChatGPT, but if employees do not see concrete examples from people like them, usage stalls. Gartner's research on digital workplace behavior has long pointed to the influence of champions and peer credibility in software uptake, and this story fits that pattern closely.
From the Encorp playbook: Social confidence often comes before process change. Teams are more likely to use AI consistently when their first training session includes live prompt practice, role-specific examples, and permission to compare results openly. That is why structured AI for Personalized Learning style programs work best when they feel practical rather than theoretical.
There is also a demographic point that many AI strategy conversations miss. Much of the market narrative still centers on developers, founders, or early-career knowledge workers. Aged Intelligence instead frames AI as support for experienced professionals carrying dense decision loads across both work and home. That widens the addressable audience for corporate AI training and changes what useful instruction looks like.
The practical lesson for mid-market teams
Mid-market companies in professional services, technology, and media can copy more from this model than might seem obvious. Not the exclusivity or the dinner format, but the mechanics: small cohorts, a narrow set of prompts, live experimentation, and examples tied to real work. Those ingredients are often more effective than broad AI courses that try to teach everything at once.
A sensible starting point is a 60- to 90-minute workshop focused on three repeatable tasks: summarising documents, drafting first-pass content, and turning messy notes into structured actions. Employees should compare outputs, discuss failures, and identify where the model introduced risk or wasted time. That makes AI training concrete. It also builds the shared language needed for later AI implementation roadmap work.
The source material offers an additional operator insight: the founders were not presenting themselves as grand technical experts. They were testing tools the way reporters do, then translating the findings into approachable language. That matters because many companies overestimate the value of expert-led monologues and underestimate the value of facilitator-led practice. In early adoption phases, confidence often spreads faster through practical interpreters than through specialists.
There is a trade-off. Peer-led formats can drift into anecdote if they are not tied to business priorities. A team may leave enthusiastic but without standards for prompt storage, tool selection, or data handling. That is where a fractional AI lead or internal sponsor becomes useful: someone needs to connect experimentation to approved workflows, budgets, and outcomes.
What larger enterprises should take from the trend
Large enterprises should read this trend less as a novelty story and more as a signal about training design. One-off AI workshops rarely stick at scale because they are detached from daily operations. What tends to work better is a cohort model: train one group, identify internal champions, collect prompts and use cases, then run the next group with sharper examples.
This is similar to how many enterprise software rollouts mature. Forrester's guidance on change and adoption programs has consistently emphasized reinforcement, role relevance, and community feedback loops over single-event training. AI is even more dependent on repetition because outputs vary and effective use improves with iteration.
There is also a content-design lesson from Berger's background at WIRED, Notion, and IDEO. Editorial instincts help because AI fluency is partly about explanation: which tasks deserve automation, which prompts produce reliable drafts, and where humans still need to check context and tone. Enterprises that treat AI training as a documentation and workflow problem, not just a procurement issue, are likely to move faster.
The main risk is fragmentation. If every team develops its own prompt library and tool habits without coordination, quality diverges quickly. Enterprises therefore need both distributed learning and central curation. That is the balance between social energy and operational discipline.
Peer-led AI learning vs traditional training programs
The market is splitting along two lines. Peer-led AI workshops generate faster trust because examples feel immediate and failures can be discussed in plain language. Traditional programs scale better when an organization needs consistency, reporting, and links to broader capability plans.
Where peer-led formats win is speed. They lower the intimidation barrier and can create momentum in weeks rather than quarters. Where formal programs still matter is standardization. Regulated processes, cross-functional rollouts, and enterprise-wide tooling all require more than informal enthusiasm.
In practice, the strongest AI training programs combine both. They borrow the intimacy of community-led learning while keeping a structured curriculum, an approved tool set, and a clear path from experimentation to implementation. That hybrid approach is what many mid-market companies have been missing: not another generic awareness session, but a repeatable bridge from curiosity to competent use.
The takeaway: AI fluency is becoming a baseline skill
The Aged Intelligence story is newsworthy because it shows AI adoption moving into a more human layer of the market: not infrastructure teams or model builders, but professionals deciding whether they can stay current without direct, practical exposure. That is a meaningful shift for any company planning corporate AI training in 2026.
What to watch next is whether more employers formalize these social learning patterns inside the business. If they do, the next wave of AI adoption may come less from new model launches and more from better-designed training environments that make regular practice feel normal.
Martin Kuvandzhiev
Co-Founder & CEO, encorp.ai
CEO and Founder of Encorp.io with expertise in AI and business transformation
LinkedIn