In partnership with
6-min read
Today: Anthropic says Sonnet 5.5 generates output faster and costs less per task. Where might that matter in your work? Also inside: AI shopping, tools for testing AI apps, expressive voices, and a calculator you can build with two prompts.
Let’s get into it 👇
Claude Sonnet 5.5 is faster—but where will you notice it?
Anthropic released Claude Sonnet 5.5, an upgrade aimed at everyday work such as fixing bugs and creating documents, slides and spreadsheets.
Image source: Claude on X
The company says it generates responses more than 30% faster than Sonnet 5 and can cost up to 30% less per task because it often uses fewer tokens. Its API token prices have not changed.
For small teams, the useful test is simple: run one task you do often with Sonnet 5.5 and compare the result, time and cost with your current model.
A routine lung scan may reveal a treatment risk
MD Anderson researchers developed AI that uses CT scans to identify patients at higher risk of lung inflammation from immunotherapy. The model showed promise in a study of 347 patients, but it has not yet been shown to improve care.
Want it weekly?
Sponsored
For product teams moving at AI speed.
AI makes it easier to ship anything, even bad ideas. The hard part is knowing which ideas are worth building.
Jira Product Discovery brings your ideas, customer insights, and priorities into one place, so your team can decide what to ship and move forward with confidence.
Capture ideas, prioritize with evidence, and build living roadmaps your team can rally around—all while staying connected to delivery in Jira, so everyone can see what’s being built and why.
Better product decisions in the AI era.
🔥Big news in AI
Image source: Noah Shinn on X
🔹Instinct and Shopify bring shopping into one conversation
Instinct says 35% of its users already use it for personal shopping. Its new Shopify partnership lets users find products with current prices and availability, check out with Shop Pay, and track orders in the same conversation. Instinct can also use preferences such as sizes and past purchases to tailor suggestions. The partnership makes shopping more seamless, though availability depends on participating merchants.
🔹Claude Code adds tools to test and improve AI apps
Anthropic introduced two Claude Code skills: build-eval helps developers create test cases from real bugs or production data, while hillclimb uses those tests to improve prompts, model choices and code. The goal is to make AI apps easier to measure and refine before changes reach users.
Sponsored
Hire anyone, anywhere — compliant in under 3 days
Found the right hire, but no entity in their country? Remote becomes the legal employer — with contracts, benefits, and tax setup handled, plus direct access to the same in-house team that runs payroll locally.
Can two prompts build a useful AI calculator?
Space Bunny Alpha is free to try, but can it make something you would actually use?
I asked it to build a simple calculator for estimating the time and value an AI subscription might save. The first prompt produced a working HTML page. A second prompt fixed the formatting and warning icons I found while testing it.
Sponsored
See what's really happening in every deal
Aligned shows you what buyers are doing between meetings inside your most important deals then surfaces the risks, openings, and next steps. No more guessing where a deal stands or finding out a champion went quiet too late. Just real visibility, so you can act before it's too late.
ElevenLabs launches v4 voice models with more control
ElevenLabs introduced two speech models: Eleven v4 adds emotional delivery, multiple speakers and sound effects, while v4 Turbo brings expressive voices to live conversations and voice agents.
Image source: Eleven v4 and Eleven v4 Turbo Text to Speech models
Creators can add directions such as “[whispers]” to a script, regenerate lines without changing the speaker’s voice, and produce speech in more than 90 languages. Both models are available through the ElevenLabs API, and v4 can be tried in its speech tool.
AI could make it harder for newcomers to stand out
A new research paper models a hiring problem: if AI makes applications less revealing about a candidate’s fit, employers may rely more on prior experience. That could make it harder for capable newcomers to get noticed. The paper is theoretical, so it does not show this is happening in real hiring.
Plus: Runway’s creative talent platform, a three-day AI film project, courses, hackathons and remote jobs.
Want it every Sunday?
New here? Get the next issue Thursday.
Join 76,000+ AI builders who read The AI Entrepreneurs every week
