Airtable Launches Superagent AI to Automate Workflows

Launching a completely new product line while your core business has lost two-thirds of its market value might seem irrational to some. However, Airtable founder and CEO Howie Liu considers it the most logical decision he could make.

During the zero-interest-rate frenzy of 2021, investors valued the company at $11.7 billion. Now, secondary markets value it at approximately $4 billion. Despite this, Airtable has raised $1.4 billion in total funding, and Liu confirms the company still holds half that amount in reserve while generating positive cash flow. The sharp decline in valuation impacted investor returns and employee stock options but didn't harm the underlying business operations.

Liu's strategic response is Superagent, an AI agent he believes could eventually surpass Airtable in significance. This marks Airtable's first standalone product in thirteen years, reflecting both the company's future direction and the current AI landscape where established software companies are competing to deliver functional AI agents.

What makes this move particularly noteworthy is Airtable's established position as a no-code platform that simplifies application development. Essentially, it's an advanced database enabling users to create custom software for their specific workflows. With over 700 employees and serving more than 500,000 organizations—including 80% of Fortune 100 companies—Airtable isn't a struggling startup but an established business making a strategic bet on a new technological architecture.

Superagent embodies Liu's vision of "multi-agent coordination"—a system where instead of a single AI assistant performing tasks sequentially, a coordinating agent manages specialized agents working simultaneously. "You're not just prompting an AI," Liu clarifies. "You're directing a team."

Here's how it operates: When you ask Superagent about expanding an athleisure brand into Europe (using Liu's example), the system first develops a research plan, identifying investigation areas and uncovering aspects you might have overlooked. It then activates specialized agents simultaneously—one examines financial considerations, another analyzes competitive landscape, while others review management and relevant news. Finally, it consolidates all findings into a comprehensive report.

The result isn't just text. It's an interactive market analysis featuring demographic breakdowns, visual competitive mapping, and customizable expansion timelines. "Imagine if everyone could access New York Times-quality data visualization for every task," Liu explained during our recent Zoom call. "This would have been unimaginable five or ten years ago when you'd only get text output. Now, receiving high-quality, interactive outputs as standard represents a fundamental shift."

Disrupt 2026 Tickets: Limited Time Offer

Tickets now available! Save up to $680 with current rates, and be among the first 500 registrants to receive 50% off your additional pass. TechCrunch Disrupt features industry leaders from Google Cloud, Netflix, Microsoft, Box, a16z, Hugging Face, and others across 250+ sessions focused on business growth and competitive advantage. Network with innovative startups and participate in curated networking opportunities that generate deals, insights, and inspiration.

Disrupt 2026 Tickets: Limited Time Offer

Tickets now available! Save up to $680 with current rates, and be among the first 500 registrants to receive 50% off your additional pass. TechCrunch Disrupt features industry leaders from Google Cloud, Netflix, Microsoft, Box, a16z, Hugging Face, and others across 250+ sessions focused on business growth and competitive advantage. Network with innovative startups and participate in curated networking opportunities that generate deals, insights, and inspiration.

Liu emphasizes technical distinctions between Superagent and competitors. He identifies Anthropic's Claude and Manus (an AI research platform being acquired by Meta) as the only other products featuring "truly capable, persistent, and intelligent agent architectures." According to Liu, most other so-called agents are essentially "LLM-powered workflows"—predetermined sequences incorporating AI calls rather than truly autonomous agents capable of self-correction and backtracking.

This distinction becomes crucial in a market saturated with AI agent announcements. OpenAI began 2025 by introducing new agent development tools, while Notion, Harvey, and numerous other companies have since incorporated agent capabilities. In this competitive environment, Liu's claims about Superagent's uniqueness must demonstrate practical value.

In his product announcement, Liu provides concrete examples of Superagent's capabilities. Request an evaluation of Google as a three-year investment opportunity, and you receive a structured analysis citing earnings calls, competitive defensibility against OpenAI and Anthropic, and overlooked risk factors. Ask for a briefing on Wells Fargo's AI strategy before a sales pitch, and you get their regulatory position, recent AI investments, and specific challenges your product could address. The system draws from premium data sources including FactSet, Crunchbase, SEC filings, and earnings transcripts.

This launch completes Airtable's transformation into what Liu describes as an "AI-native platform." Last autumn, the company appointed David Azose, former engineering lead for ChatGPT's business products at OpenAI, as CTO. Simultaneously, it acquired DeepSky (previously Gradient), an AI agents startup that had raised $40 million. Superagent will operate with considerable autonomy under the leadership of DeepSky's founding team.

Pricing details were still being finalized last week, but indications suggest standard AI product pricing: $20 monthly per user for entry-level access, scaling to $200 for power users, including generous inference credits. "We're not currently focused on profit margin optimization," Liu states.

Whether Superagent achieves Liu's vision of a trillion-dollar market or becomes an ambitious gamble remains uncertain. Competition is substantial, and the technical distinctions Liu emphasizes between "real agents" and alternatives may not concern customers if competitors deliver satisfactory results more quickly and affordably.

For a CEO whose company has experienced a $7.7 billion valuation decrease while maintaining most operational capital, this move demonstrates commitment to future growth over protecting current status. Liu has reframed the valuation reduction as a recruitment advantage, telling employees they're obtaining "more attractively priced equity compared to the $11 billion valuation" with significant potential upside. The company has sufficient capital for strategic acquisitions and doesn't require additional funding rounds.

When asked if Superagent represents the greater long-term opportunity, Liu acknowledges the possibility. Airtable "will likely remain larger than any new products, including Superagent, in the immediate future," he notes. "But I appreciate having the opportunity to invest in Superagent. Maintaining options is valuable."

This approach represents what Liu terms "wartime" leadership—a concept he initially rejected as unnecessarily aggressive but now considers appropriate. "The ability to adapt quickly," he observes, "is currently the most value-creative operational approach." He adds, "It's also the most exciting way to work."