The news
Startups are betting that AI agents in messaging will replace many mobile apps. On October 1, 2026, TechCrunch reported that Photon, which helps developers build agents that work inside messaging apps, raised a $4.5 million seed round. Days earlier, Wabi pivoted to a messaging product, and Instinct brought its agent into group chats.
Photon offers a single API (application programming interface), a framework for adding channels, a command-line tool and monitoring for agents that run on iMessage, WhatsApp, Telegram, SMS and RCS, email and voice. Its round was co-led by Gradient and A*, with Vercel, HSG (HongShan), Z Fellows and others joining, according to TechCrunch. The company reported more than 40,000 developer sign-ups, 10x revenue growth in four months and churn under 3%. On September 17, it staged an app funeral in San Francisco.
Wabi, founded by Replika creator Eugenia Kuyda, announced Wabi 2.0 in late September as an invite-only release, TechCrunch reported on September 29. The product moves from a prompt-based app builder to a messaging platform that combines chat, app creation and task management, generating custom interfaces on demand, such as a calorie tracker or a family activity calendar. Kuyda argued that chat-only agents bury history and ongoing tasks in one endless scroll.
Instinct, founded by Noah Shinn and valued at $10 billion after its Series C, said on October 5 that users can add its agent to group conversations to plan trips, buy event tickets or coordinate carpools, as reported by TechCrunch. Friends without Instinct accounts can take part. Personal agents ask permission before connecting to a group agent or sharing information, and users can revoke a group's access at any time. The feature starts with early-access users.
A TechCrunch roundup published October 3 counted a crowded field of text-based agents. It included Poke, which TechCrunch says was the first AI agent approved for Apple Messages for Business in June 2026, and Town, which raised a $55 million Series A that month.
The numbers
- Photon seed round
- $4.5 million
- Photon developer sign-ups
- 40,000+
- Photon revenue growth
- 10x in four months (company figure)
- Instinct valuation
- $10 billion
- Wabi pre-seed (Nov 2025)
- $20 million
Why CEOs should care
For chief marketing and digital officers, the shift raises a practical question: how does your brand get found and transact when a customer asks an agent in a text thread instead of opening your app? Audit whether your products, prices, inventory and booking flows are available through APIs an agent can call, and whether you have a presence on business messaging channels such as Apple Messages for Business or WhatsApp.
For CFOs and product leaders, the bet cuts against years of app investment. Before funding the next app rebuild, ask what share of customer tasks could be completed through messaging and agents, and what it would cost to support those channels.
For CISOs and privacy teams, Instinct's group-chat design shows the controls customers will expect: explicit permission before an agent shares data or acts, isolation between personal and group agents, and easy revocation. Any agent integration you build should meet that bar, with logs of what each agent did on whose behalf.
The bigger picture
The common thread is that messaging is where people already spend time, so agents that live there avoid the friction of downloads and logins. Photon CEO Daniel Tian told TechCrunch the move from apps to agents will take a while but is, directionally, inevitable. Large platforms are in the race too: TechCrunch names Meta's Muse and OpenAI's ChatGPT Dots as Instinct's competitors.
Company growth figures, such as Photon's 10x revenue jump, are self-reported, and none of the startups disclosed revenue totals. The direction is clearer than the scale.
What’s next
Watch how quickly Instinct moves its group-chat feature beyond early access, whether Wabi 2.0 opens past invite-only, and whether Apple and Meta approve more third-party agents on their business messaging channels, which would decide how easily brands can plug in.
What “Fact-checked” means
Fact-checking means testing a story’s facts against the evidence before it is published. This story went through at least two separate checks before this version was published.
- What we checked
- Its names, figures, dates, job titles, quotes and who said what were checked against the story’s sources, including its main source where it could be opened. The headline was checked for accuracy and overstatement.
- How
- A first check reviewed the whole story. If it passed, a second, skeptical check went back to the sources to look for mistakes in the most important facts. If a check flagged the story, it was edited to fix the problems found, and a separate re-check then reviewed the whole story again.
- Who
- The checks are made by our newsroom, as steps kept separate from the writing, under rules set by our editor, Hussein Mukhtar. A story the checks still flag is held for the editor, who decides whether it is fixed, published or dropped.
- If something is wrong
- “Fact-checked” does not mean error-free. If a material error is found after publication, we correct the story and add a note saying what changed. Report an error
