UnitedHealth breaks down how it plans to spend $1.5B on AI
THE SO WHAT
Healthcare IT just got a new benchmark — $1.5B in AI spend from a single payer means every major insurer and provider is now under pressure to turn AI from slideware into operating margin and sellable product. If you’re in healthtech and not explicitly building for payer workflows and data exhaust, you’re walking away from the only customers with this scale of budget.
READ THE SOURCE
MORE FROM THE WIRE
Marketing to "a segment of one" is possible with AI, says Adobe's enterprise CMO
“Segment of one” is just table stakes once you have real-time behavioral data and AI scoring—your constraint is governance and content ops, not modeling. If you’re still running quarterly audience definitions, your marketing org is structurally misaligned with what the tooling can already do.
Applied AIOpenAI says ChatGPT Images 2.0 comes in Instant and Thinking variants and can generate images of up to 2K resolution and in multiple aspect ratios (Zac Hall/9to5Mac)
Splitting Images 2.0 into Instant vs Thinking formalizes a tradeoff every product team is already making—latency vs reasoning. If your UX doesn’t expose that choice to users, you’re leaving performance or quality on the table in every creative workflow you ship.
Applied AIOpenAI says that ChatGPT Images 2.0 has a stronger understanding of non-Latin text rendering in languages like Japanese, Korean, Hindi, and Bengali (Amanda Silberling/TechCrunch)
Non‑Latin text that actually renders correctly in images is the unlock for using generative visuals in global markets—menus, signage, packaging, local ads. If your brand or product is multi-region and you’re still designing English-first then localizing, your workflow is already obsolete.
Applied AIOpenAI Unveils New Image Model That’s Better at Charts and Diagrams
Accurate charts and scientific diagrams move image models from marketing toys into analyst and R&D workflows—where a mislabeled axis is a real liability. If your teams are still hand‑building every visualization, start reallocating that time to checking model output and refining prompts instead.