HeyJaredAddition
Insights and experiments at the intersection of Generative AI and marketing.
- Indexed issues, last 90 days
- 3
- Latest publication
- Sep 23, 2026
- Audience
- Checking…
- Earliest in this view
- Jul 23, 2026
Latest issues
The shape of creative AI systems (opens the original)
Read excerpt
Most of the creative AI systems we’ve built can be understood as three interacting kinds of logic.Creative logic is the expressive intent: brand guidelines, visual references, tone, aesthetic rules, examples, constraints, and all the domain-specific judgments about what good looks like.Thanks for reading! Subscribe for free to receive new posts and support my work.<input class="e
Machine writable brands (opens the original)
Read excerpt
Back in 2022, we participated in an “ad Turing test” run at an early AI marketing conference. The challenge was to build an AI system that could produce finished ads entirely on its own. Most of the contestants created HTML templates in ChatGPT that they slotted copy and images into with LLMs and image models. We attempted to take a different approach, interpreting the contest’s “fully generative” rule to mean that even the layout needed to be generated at inference time by a model. We set up ou
The Cannes Rift: Why We're Not Picking a Side (opens the original)
Read excerpt
Every year, Cannes becomes the forum for debates that have been stirring long before attendees hit the fiery hot Croisette. This year it was AI versus craft. Technology-heavy, AI-forward work on one side. Traditional, human-driven creative on the other. It showed up in the sessions, in the parties, in the presentations — and honestly, it showed up in a lot of the work itself, which tended to fall cleanly into one camp or the other.We spent the week increasingly convinced that this is the wrong d
Publishing over time
Last 90 days. Choose a month to open its work.
Recurring subjects
Named in the text we hold. One piece can cover several.
Audience
No verified audience measurement yet.
About this data
Counts cover the work we have indexed. Tone needs enough text and a confident classification. Excerpts and episode notes are not full articles or transcripts.
Identity or attribution wrong? Suggest a correction.