What Are A.I. Billboards in San Francisco Trying to Tell Us?
If you’ve driven through San Francisco lately, you’ve probably noticed something strange. Not the traffic. Not the self-driving cars. Not even the rent. The billboards. More specifically, the AI billboards. They’re everywhere.
Towering above Highway 101, lining Interstate 80, peering over downtown rooftops like digital prophets trying to deliver an important message before your next exit. The problem is that nobody seems entirely sure what that message is.
One billboard simply says: “Agents.” That’s it. Agents. No explanation. No context. No indication whether they’re software, spies, real estate brokers, or characters from a science fiction movie.
Another declares: “The Future Is Autonomous.” Wonderful. Autonomous what? Cars? Toasters? Tax accountants? My now 21-year-old kid is still living at home? Then there are the truly mysterious ones: Reasoning. Inference. Multi-Modal. Scale.” These aren’t advertising slogans so much as vocabulary words.
Imagine if Coca-Cola’s billboard simply said: “Carbonation.”Or if Nike went with:”Footwear.”
Yet somehow Silicon Valley has convinced itself that cryptic technical nouns are compelling marketing. Perhaps these billboards aren’t meant for ordinary humans. Maybe they’re designed specifically for venture capitalists commuting from San Francisco to Palo Alto. The rest of us are merely accidental viewers. What an expensive way to get VC investors’ attention. Well, probably not. Many of these new companies, even without products, are already valued at billions, so throwing out twenty to fifty thousand to say “we exist, even though we aren’t making a product yet” is rather cheap. This would explain a lot.
However, if these signs were intended for the public, they’d probably answer basic questions such as:
- What do you do?
- Why should I care?
- How will this improve my life?
- Are you replacing my job?
Instead, the messaging often feels like it was written by engineers who were told they had exactly three words and absolutely no verbs. The verbs just get in the way. We need clear statements. One company proudly announces that it is building the infrastructure layer for agentic orchestration. Another promises enterprise-grade foundation models. By the way, who determines what’s “enterprise grade”? A third appears to be advertising a concept rather than a product.
It’s entirely possible that every AI billboard is actually saying: Dear Investors: We have a large language model. Please call us before our next funding round.
At 65 miles per hour, these distinctions become somewhat difficult to appreciate. Of course, there is something oddly fitting about all of this. For years, technology companies sold us products. Then they sold us platforms. Then ecosystems. Now they’re selling us possibilities.
The AI billboard has become less of an advertisement and more of a philosophical statement. It’s less “buy this” and more “contemplate this.” Perhaps the billboards are intentionally vague because nobody—including the companies themselves—is entirely sure where AI is heading.

Maybe the signs are not marketing at all. Maybe they’re in collective therapy. A giant public brainstorming session stretched across Northern California’s freeway network, which only requires the right clues to pull the meaning all together.
Or perhaps the message is surprisingly simple: “We are an AI company.”
“Yes, us too.”
“And us.”
“And us.”
“And definitely us.”
In a region where every startup is racing to become the next transformative technology company, the billboards have become the modern equivalent of planting a flag on a hilltop. Whether drivers understand them may be beside the point. The real audience may be everyone else building AI. The billboards aren’t speaking to us. They’re speaking to each other.
And somewhere between San Francisco and San Jose, an entire conversation is taking place in giant sans-serif type that most commuters can only partially understand. Agents. Inference. Reasoning. Scale.
The rest of us just keep driving, nodding thoughtfully, and pretending we know what they mean, or not.