In Meta's Version of the Future, My Muse Will Call Yours
"... everyone is just going to have to adapt," Meta chief financial officer Susan Li said during the third-quarter earnings call.
By Ellen Thomas | October 22, 2026, 6:15 PM UTC
A Future Where AI Agents Do the Talking
Meta is sketching out a future in which AI agents operate on your behalf—and not just as passive assistants waiting for instructions. The company's vision, articulated on its most recent earnings call, centers on a concept it calls "Muse": a personalized AI agent that could eventually interact directly with other people's agents.
The headline scenario is simple to state and radical in its implications: I'll have my Muse call your Muse. In that model, the tedious back-and-forth of scheduling, negotiation, information exchange, and coordination would happen between AI agents rather than between people.
'Everyone Is Just Going to Have to Adapt'
Meta CFO Susan Li addressed the shift during the company's third-quarter earnings call, acknowledging that this agent-mediated future will require adjustment across the board. Her remark—that "everyone is just going to have to adapt"—captures the company's position that the transition is not optional.
The comment reflects a broader 2026 trend in which major technology platforms are moving from single-turn AI assistants toward persistent, personalized agents that hold context, act autonomously, and interoperate with other agents. Meta's "Muse" framing places that concept at the center of its product roadmap.
What Agent-to-Agent Interaction Means
If personal agents can talk to one another, several everyday workflows change fundamentally:
- Scheduling and coordination: Calendars and availability could be reconciled automatically between two people's agents, eliminating email threads.
- Information exchange: Agents could share documents, preferences, and context directly, with the human users setting the rules of engagement.
- Negotiation and transactions: Routine purchasing or planning decisions could be delegated to agents operating within user-defined constraints.
Each of these scenarios raises the same underlying questions: how much autonomy users grant their agents, how agent-to-agent exchanges are authenticated, and who is accountable when an agent acts on a user's behalf.
The Stakes for Meta—and Everyone Else
Meta's bet on personal agents fits its broader 2026 strategy of embedding AI across its family of apps and devices. Framing the future around agent-to-agent communication signals that the company sees interoperability and persistent personalization—not just model capability—as the next competitive frontier.
For users, the message is that adaptation is coming regardless of preference. For the industry, Meta's public positioning sets an expectation: the next phase of AI will be defined less by what a single assistant can answer and more by what a network of personal agents can accomplish together.
As Li put it, everyone is just going to have to adapt. Whether that adaptation arrives smoothly or otherwise, Meta is already planning for the moment when your Muse and mine start talking.
via The Verge AI