We scan new podcasts and send you the top 5 insights daily.
Ramesh says customers keep telling her that out-of-box image generators make everything look the same. She argues millions of businesses using one generic model cannot all be solved, so understanding taste and adding human preferences is the next design problem Coco Labs is chasing, rather than execution alone.
AI won't replace designers because it lacks taste and subjective opinion. Instead, as AI gets better at generating highly opinionated (though not perfect) designs, it will serve as a powerful exploration tool. This plants more flags in the option space, allowing human designers to react, curate, and push the most promising directions further, amplifying their strategic role.
AI tools struggle in creative processes because they cannot "see" or have personal preferences. Their output is limited by the user's ability to verbally describe visual inspiration, creating a significant bottleneck. This highlights why human taste and curation remain essential for high-quality creative work.
Users are dissatisfied with purely AI-generated creative outputs like interior design, calling it "slop." This creates an opportunity for platforms that blend AI's efficiency with a human's taste and curation, for which consumers are willing to pay a premium.
Frontier AI models struggle to produce differentiated, high-end creative work when guided purely by open-ended prompts. Providing concrete aesthetic references prevents output from appearing low-budget or generic, ensuring the human directs taste while the model handles execution.
Services like Taste Labs, much like Squarespace, aim to democratize good design. However, this commoditization often leads to a recognizable, copied aesthetic. As AI models are trained on what is deemed "tasteful," they may converge on a single style, undermining the originality that is a key component of true taste.
Rather than optimizing solely for performance on standard industry benchmarks, Ideogram focuses on embedding a subjective quality of "taste" into its models. This requires using human designers for evaluation, as they believe current AI is poor at judging aesthetic nuances, giving them a unique creative edge.
As AI handles implementation, the new limiting factor is the user's "taste" and deep domain knowledge. To get great subjective output, like design, the user must provide high-quality references and have the expertise to judge the AI's work, knowing when to push for better results.
Anusha Ramesh of Coco Labs says image generators act like photographers and produce flat pictures, while its Vibe model builds designs in layers. That lets users keep custom fonts, colors and brand style editable, rather than a single image that generic models flatten into sameness.
Despite AI's ability to generate functional code, replicating the nuanced, subjective quality of a specific designer's "taste" remains extremely difficult. Felix Lee, after spending weeks attempting to codify his own taste into an AI model with little success, notes it's a significant unsolved challenge.
According to Dreamer's CEO, the biggest capability missing from LLMs is "taste." By default, AI-generated applications and UIs are generic and identifiable by the model that created them. It requires extensive human effort in prompt engineering and templating to create delightful, non-generic user experiences.