Tastewise

Tastewise provides a verticalized intelligence layer for the food and beverage industry. Its current product combines a large food knowledge graph with consumer panels, social and discussion signals, restaurant menus, recipes, retail and e-commerce observations, and customer first-party data. The company says its system structures those sources with food-specific taxonomies and models, refreshes the underlying signals frequently, and exposes the result through dashboards, reports, TasteGPT, APIs, and configurable AI agents. The important technical distinction is not simply access to a general-purpose language model: it is the normalization, entity mapping, category context, and evidence retrieval needed to answer questions about ingredients, occasions, claims, menus, brands, and demand without losing the underlying source trail.

The target customer is an enterprise food or beverage team whose decisions span product development, brand strategy, consumer research, sales enablement, and foodservice or retail execution. A CPG innovation team can use the platform to identify whitespace, compare ingredient or flavor momentum, validate a concept, and create a formulation or marketing brief. A sales organization can turn category and shopper evidence into a retailer-specific growth narrative. Restaurant and foodservice operators can examine menu patterns, occasions, and competitive positioning. Tastewise is increasingly packaging these workflows as agents that can run inside the company platform or connect to tools such as ChatGPT, Claude, Microsoft Teams, Slack, Outlook, and APIs. That makes the commercial proposition more actionable than a periodic trend report, while increasing the importance of permissioning, auditability, and output quality.

The market opportunity is substantial but crowded. Tastewise competes with established market-research and syndicated-data providers such as Mintel, Innova Market Insights, Euromonitor, and Datassential, as well as food-specific AI and trend platforms such as Spoonshot. It also competes indirectly with internal insights teams, agencies, retailer data products, social-listening tools, and general AI assistants connected to customer data. Tastewise's defensibility therefore depends on the depth and freshness of its food ontology, proprietary labeling and evaluation, cross-source coverage, workflow integrations, and the accumulated behavioral context of enterprise customers. A model demo is easy to copy; reliable evidence-backed answers across markets, categories, and commercial workflows are harder. The company’s own site reports broad brand adoption and large data volumes, but those are company claims that require diligence against contracted customers, retention, usage, and independently measured outcomes.

Commercial evidence is stronger than the previous early-stage classification suggested. Tastewise was founded in 2017, announced a $17 million Series A in 2022, and announced a $50 million Series B in June 2025 led by TELUS Global Ventures, with additional participation reported from Disruptive AI, PeakBridge Capital, Duo Partners, and PICO Venture Partners. Public company materials position the business as serving major global food brands and expanding its AI-powered marketing, sales, and innovation workflows. This supports a mid-stage private startup classification, but it does not establish profitability, durable net retention, or product-market fit across all claimed segments. Key diligence questions are revenue quality, concentration, gross margin after data acquisition and model inference costs, the conversion of pilots into recurring deployments, and whether agent usage produces measurable innovation, distribution, or sales outcomes.

National-security relevance is limited and indirect. The same capabilities could help a government, humanitarian organization, or military logistics contractor monitor food availability, local consumption patterns, substitution options, or provisioning demand during a crisis. However, Tastewise is not a defense, intelligence, supply-chain security, or critical-infrastructure company, and public evidence does not show defense customers, government contracts, classified deployments, or mission-specific assurance. Its commercial data sources may be biased, delayed, restricted by platform terms, or unsuitable for high-consequence decisions. Any security application would require new controls for provenance, privacy, access, resilience, offline operation, uncertainty reporting, and procurement compliance. The appropriate database conclusion is weak adjacency rather than substantive dual use.