A global healthcare company
Healthcare and life sciences, brand and marketing
A generative AI introduction for a brand and marketing team in June 2023, built around what the technology could and could not do for their work, rather than around the tools of the month.
The problem
The brand and marketing team wanted to know what generative AI could do for its work, and how to start. This was June 2023, when almost every answer available publicly was either a demo or a warning.
What I did
I prepared and delivered the session on 13 June 2023, contracted through the institute rather than personally.
It covered what foundation models actually are, and how they differ from the task-specific models the team had already met in analytics; prompting against tuning, and when each is the right instrument; the cost trade-off between building a foundation model and using someone else’s; and what image models can and cannot do, which for a brand team is the question with the sharpest edges.
Examples used GPT-4 vision, ChatGPT, DALL-E 2 and Midjourney, because in a session like this a live example settles an argument that a slide cannot.
What came of it
No outcome was measured. It was a single session, and I have no evidence of what the team did next.
Other training work
Have a problem worth measuring?
I take a small number of advisory and build engagements each year.