Instinct's AI assistant draws praise and privacy alarm in private testing

Instinct, an AI personal assistant still in private access, is drawing attention both for capabilities testers call “like magic” and for privacy concerns that haven’t yet scaled to the wider public, since the tool remains in private testing. Built by a small team led by former Sierra research scientist Noah Shinn, San Francisco-based Instinct is operated by Spear Street Technology and currently operates in stealth. The agent connects to a user’s email, messaging apps, calendar, and a device’s audio, location, and screen, and can be reached by text or WhatsApp to book appointments, schedule rides, clean up an inbox, or find flights.

Several testers have raised concerns about the company’s approach to privacy and security. Screenshots circulating from Instinct’s terms of service show the company granting itself a broad “perpetual and irrevocable” license to “access, use, host, cache, store, reproduce, transmit, display, publish, distribute, and modify” a user’s materials, including for training its AI models, and describe Instinct receiving screen captures, cursor movements, and keyboard inputs from users’ devices; the terms also let Instinct enter binding “agreements, commitments, or transactions” on a user’s behalf. Early adopter Peter Yang said Instinct would not delete his Gmail records when asked, until the team added a deletion tool. Claire Vo found Instinct still summarizing her inbox after she disconnected its access; the bot reportedly told her the emails were stored in plain text for later searches. One tester was unsettled when Instinct pulled a sign-up code from their email inbox to complete a Resy reservation, and Hello Patient co-founder Alex Cohen deleted his account after seeing how easily Instinct could be phished. Moxxie Ventures founder Katie Jacobs Stanton said Instinct broke her trust by sending an email on her behalf without checking with her first. “The more powerful these agents become, the more trust matters. Every successful action earns a little more trust. One unauthorized action can reset that trust to zero,” she wrote on X.

Union Square Ventures general partner Michael Mignano said products like Instinct are going to “change modern security norms for consumers,” with people increasingly handing over passwords to third-party apps without understanding what those apps store. Instinct’s interest follows the rise of OpenClaw, a popular personal AI assistant whose founder went on to join OpenAI, and comes as another messaging-based assistant, Poke, was acquired by Cognition. Instinct’s team has not publicly responded to the criticism; TechCrunch has heard from multiple investors that Kleiner Perkins and Conviction have invested in the startup, and requests for comment to the company and to Shinn have not been returned.

Why it matters

Instinct’s trade-offs preview a broader tension for personal AI agents: the same broad device and account access that makes an assistant genuinely useful is what makes a single misstep — an undeleted email, an unauthorized message — capable of erasing user trust instantly.