‘It’s just his AI and my AI going back and forth’: The workplace phenomenon that’s undermining human relationships

The article highlights “socially offloading” at work—employees and managers using AI to script or interpret boss-employee communications—arguing it can erode human skills like emotional intelligence and relationship-building. It links the trend to leadership vacuum effects from flatter orgs and reduced middle-management coaching, citing Meta’s job cuts of 25,000 since 2022 and its AI-heavy management model. Overall, it’s a cautionary view that AI adoption may help efficiency, but raises cultural/execution risks if training and mentorship don’t keep pace.

Analysis

The real economic signal is not “AI replaces communication,” but that the highest-value layer in the enterprise may shift from drafting to judgment translation. That makes SKIL the cleaner beneficiary only if customers start paying for measurable behavior change — practice, coaching, and manager enablement — rather than another generic AI wrapper. If that spending appears, it is more durable than one-off productivity software because it ties to attrition, promotion quality, and manager effectiveness, which enterprises defend in softer labor markets.

For META, the near-term market still rewards organizational compression, but the second-order risk is execution quality: fewer managers can improve margin optics while quietly degrading talent development and team cohesion. That risk usually shows up with a lag of 2-4 quarters through slower product iteration, higher regrettable turnover, or more internal rework, so it is not a day-trade short. The bullish case on efficiency remains intact unless those soft metrics start to leak into guidance.

Contrarian view: consensus likely overestimates immediate AI productivity and underestimates the cost of humans outsourcing conflict, coaching, and negotiation. This is more a structural productivity drag on weak organizations than a sector-wide earnings shock, so the tradeable edge is limited until HR spend or churn data confirms it. The biggest risk to the thesis is that firms simply codify AI-assisted communication and preserve output without visible damage to quality.

Key Decisions for Investors

• SKIL: keep on a 1-3 month watchlist for a tactical long only if management commentary shows enterprise demand for coaching/role-play products translating into bookings or renewal-rate improvement. Target is a rerating if investors start treating it as AI workflow software rather than legacy edtech; thesis fails if net retention stays soft.
• META: do not short on this article alone. Use it as a medium-term monitoring item for execution risk; the bearish view only becomes actionable if next 1-2 quarters show rising attrition, slowing product velocity, or guidance that reflects coordination friction. Otherwise the margin story remains dominant.
• No immediate trade in AERA, GAP, HRDI, or TCHC from this signal alone. The relevant catalyst is downstream HR/training budget allocation and employee churn, which is a 6-18 month data set rather than a headline catalyst.
• If you want expression, prefer a relative long SKIL versus a broad software basket only after evidence of enterprise adoption; that is cleaner than a direct short on AI-efficiency winners, which the market can keep rewarding for months before any cultural drag appears.