The buzzy artificial intelligence startup Instinct has just revealed a new potential revenue stream centered around product recommendations, but the rollout is already rubbing a significant portion of its user base the wrong way.
Overnight, the company’s AI agent began pushing unsolicited product suggestions to users as part of a newly introduced feature called Instinct Selections. The feature is designed as an effort to curate personalized lists for users across various lifestyle sectors, including dining, travel, and shopping.
Founder Noah Shinn announced the feature’s debut on social media, explaining that the foundational idea behind the rollout is to partner with local chefs, designers, architects, travel guides, and other industry professionals to bring what he described as "human taste" directly to the AI’s product offerings. In practice, this means that instead of solely relying on what the automated system could suggest based strictly on its training data gathered from the wider web, the company is partnering with real-world experts to offer human-curated suggestions alongside automated insights. However, the startup has not yet publicly shared the specific identities of its human partners.
Detailing the vision behind the new capability, Shinn wrote on X that the next time a user is looking for a restaurant, Instinct can pull from a handpicked list curated by professional chefs who know the hidden gems in a given local area. Furthermore, he noted that if a user wants to find a new hiking trail to explore, the AI agent can draw on recommendations from local guides to suggest a few specific routes tailored to individual preferences. For users looking to add color and character to their living spaces, the platform aims to find home decor pieces from independent designers that match the user’s personal style, available space, and budget constraints.
The stated goal of the initiative is to deliver world-class recommendations that stand out in quality from what everyday consumers might typically find on the broader internet or through competing chatbots. Yet, despite the lofty ambitions of the founding team, the initial reaction to these seemingly unsolicited product pushes has been far from universally positive.
Numerous early users took to public platforms to voice their frustration over receiving unexpected commercial suggestions. Shruti Gandhi, a general partner at the AI-focused venture capital firm Array VC, expressed sharp disappointment after waking up to commercial prompts. She pointed out that she had no underlying need for carry-on luggage, sunglasses, or a brim hat, criticizing the new direction by comparing the experience unfavorably to a clumsy digital storefront.
Other prominent tech figures echoed similar sentiments. Andrew Yeung, an entrepreneur, confirmed on X that he also received a series of product recommendations that he had never requested, describing the experience as decidedly off-putting.
Another user, Chat Joglekar, who serves as the founder of a business-buying marketplace, detailed his own discomfort with the new feature. He described having his first uncomfortable moment with an AI assistant when Instinct began aggressively suggesting products that he should purchase based on sensitive personal data gleaned from his email correspondence and upcoming travel plans. Joglekar noted that the system suggested a to-go coffee flask simply because he shared a recurring coffee subscription, alongside various travel accessories derived from private itinerary notifications.
The overarching challenge facing developers of consumer-facing AI agents is that the utility of curated suggestions relies heavily on context. To feel genuinely helpful rather than invasive, recommendations must either be explicitly requested by the user or fit naturally within the immediate conversational flow. When product suggestions arrive at random, unprompted intervals—as Instinct’s early implementation has demonstrated—they run the severe risk of being perceived as unwanted spam rather than intelligent assistance.
This misstep is unlikely to endear the startup to users at a time when the market is becoming increasingly crowded. Instinct’s agent is competing directly against powerful, well-established AI agents from much larger technology giants like Meta and OpenAI, which currently refrain from pushing unsolicited commercial products into their core chat experiences.
While Instinct has not explicitly confirmed whether it is currently generating direct revenue from these product recommendations or what its formal monetization roadmap entails, the feature is fundamentally structured for commerce, advertising, or affiliate partnerships. TechCrunch has reached out to the company for additional comment regarding the mechanics and future plans of the new feature.
The controversial product rollout follows closely on the heels of major financial milestones for the company. Instinct recently confirmed a massive $1 billion Series C funding round, which places the startup at a staggering $10 billion valuation. Although the company has not yet disclosed official figures regarding its active user base—metrics that would typically help justify such a high valuation—founder Noah Shinn recently shared key economic insights during a podcast appearance with investor Patrick O’Shaughnessy.
During that interview, Shinn revealed that the platform is rapidly approaching a milestone of $1 billion in annual transactions processed directly through the AI agent. He added that more than half of all transactions handled on the platform are related to travel, underscoring the heavy transactional utility the agent already provides to its consumer base even as it navigates early stumbles in its commerce strategy.
