People covered the business side. 130-word statement. Dot plot shifted hawkish. MGM below bid. PCE forecast from 2.7% to 3.6%.
Here's what it means for the technology org.
Three Constraints, Simultaneous
Warsh's first FOMC just compressed People Inc's technology investment window from three directions at once.
Constraint 1: The Budget Math Gets Worse
In The ARPU Architecture, I calculated $150M free cash flow minus $122M annual debt service leaves $28M for everything — platform development, AI integration, the entire Inversion build. That was at current rates.
Now the median dot projects 3.8% year-end, up from 3.4% in March. Nine of eighteen FOMC members project at least one rate hike before December. People Inc carries $1.9B in debt. Every 25 basis points of rate increase on refinancing adds roughly $4.75M in annual debt service.
One hike compresses the residual to ~$23M. Two hikes to ~$18M.
Translation: The technology org may need to cut 15–35% of its already-constrained investment budget before a single line of code is written. The six systems I've recommended across 32 posts don't all survive this math. Maybe three do.
If the MGM deal closes — and at $47.75, below the $48.30 bid, the market is pricing some doubt — the combined entity's debt load grows substantially. People's analysis puts the incremental interest cost at $45M per 25bp. In a rate-hike environment, that's capital that doesn't flow to technology.
Constraint 2: The Regulatory Surface Expands
Warsh announced five task forces. One examines “productivity and jobs in an era of transformation” — surveying the pace, reach, and economic impact of “new general-purpose technologies, including AI.”
This matters for People Inc specifically:
- D/Cipher's OpenAI dependency. The Vendor Problem identified OpenAI as a single-vendor risk for content understanding. A Fed task force studying AI's labor impact adds regulatory surface to that dependency. If the task force's findings shape policy around AI in content operations, D/Cipher's embedding pipeline sits directly in the crosshairs.
- The 50% productivity claim. In The Productivity Question, I examined the industry pattern of claiming AI drives 50% content productivity gains. Condé Nast, Future plc, and People Inc have all made versions of this claim. A Fed task force asking “does AI actually boost productivity or just displacement?” creates discomfort for companies whose investment thesis depends on the answer being yes.
- The AI content pipeline. If you're generating or augmenting content with AI at scale — recipe variations, product descriptions, SEO-optimized rewrites — a federal productivity review adds a new consideration to the build-vs-buy calculus. Not regulation yet. But the signal that regulation could follow.
The task force reports by year-end. That's the same timeline as People Inc's fiscal planning cycle.
Constraint 3: The Competitive Gap Widens
In five days, the ad tech industry gathers at Cannes Lions. The theme this year isn't generative AI as a creative toy. It's agentic AI as marketing infrastructure.
What's shipping:
In The Peer, I mapped this gap: Hearst's AURA IQ is shipping agentic capabilities on AWS infrastructure while D/Cipher remains a contextual targeting engine without agent interfaces. PubMatic's Decision Fabric now makes the gap concrete — partner AI models run inside the programmatic auction, not as a separate contextual signal passed upstream.
D/Cipher can't participate in this architecture because it wasn't built for it. It generates content embeddings. Those embeddings inform targeting segments. But it has no agent interface, no MCP endpoint, no ARTF container. It's a content intelligence layer without a programmatic nervous system.
This is the constraint that bites hardest in the squeeze: the investment needed to close this gap just got more expensive, while the gap itself just got wider.
What Survives
Three priorities for a compressed budget:
1. D/Cipher → ARTF adapter. Not a rebuild. A protocol adapter that exposes D/Cipher's contextual signals as an ARTF-compatible container. Cost: small team, 2–3 months. This is the minimum viable agentic capability. Without it, D/Cipher becomes invisible to the programmatic ecosystem that's forming at Cannes.
2. Graphene-Selene as cost defense. The Recipe Graph argument holds: structured data in a knowledge graph serves multiple consumers (website, D/Cipher, AI agents, product apps) from a single investment. In a squeeze, the platform that reduces marginal cost per consumer is the one that gets funded.
3. OpenAI diversification. Not because the licensing revenue ($16M/year) is at risk. Because the Warsh task force adds regulatory uncertainty to any AI dependency. Evaluate alternatives for the embedding pipeline — not to replace OpenAI, but to have a credible Plan B if the regulatory climate shifts. DDM already partners with multiple AI companies; extending that to the D/Cipher pipeline is insurance, not panic.
The Clock
The squeeze has a timeline:
June 22–26: Cannes. The agentic advertising standard solidifies without D/Cipher.
Summer: Warsh task forces begin work. AI productivity narrative under scrutiny.
Fall: Task force initial findings. People Inc fiscal planning cycle.
Year-end: Rate decision. Dot plot resolution. Task force conclusions. Budget finalized.
Six months. The window to make three technology decisions — agentic adapter, knowledge graph investment, AI vendor diversification — while the budget allows it.
The 130-word statement was short on purpose. What it didn't say was louder than what it did: no forward guidance, no easing bias, no reassurance. The technology org should read the silence the same way the bond market did.
Plan for higher rates. Build for less capital. Ship what matters.