Wired tests six AI agents, ranks Gemini Spark top for news briefings
A software reporter at WIRED set up six always‑on AI agents on an iPhone—Meta’s Muse, OpenAI’s Dots, Instinct, Anthropic’s Claude Duet, Google’s Gemini Spark, and xAI’s Grok Bot—to see how they handle everyday tasks. Prices ranged from free (Muse) to $100 a month (Dots), and each agent required access to personal data such as email and calendar.
Key points
- Six AI agents—Muse, Dots, Instinct, Claude Duet, Gemini Spark, Grok Bot—were tested on an iPhone
- Gemini Spark scored 5/5 for news briefing; Dots scored 5/5 for schedule conflict and lunch ordering
- Pricing spans free to $100 per month, and agents require access to personal email and calendar data
The author ran three core tests over 24 hours: generating a morning news briefing, ordering lunch for a small team, and detecting a scheduling conflict between a calendar event and an email. Gemini Spark earned a perfect 5/5 for the news brief, producing a headline that highlighted a retail‑agent clash. Dots also scored 5/5 for both the schedule‑conflict alert and the lunch order, correctly flagging the overlap and placing a $30.57 order through DoorDash. Other agents performed inconsistently, with Muse and Claude Duet scoring mid‑range and Instinct, Grok Bot and Gemini Spark on lunch ordering receiving low marks.
The piece notes that while agents are improving—especially in browser control and task follow‑through—they still raise privacy and security concerns, as they store all interactions in a single thread and can inadvertently expose sensitive information.
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Agents & Tools
All →- OpenAI’s dots score 8.8 in hands-on test, rank third among five agents · 1 src
- Developer builds voice-driven AI agent with ledger, multiple PCs and Claude Code · 1 src
- IBM adds governed identity and credential controls for enterprise AI agents · 1 src
- LiteLLM 1.107.0 compromised, developer says release process was hijacked · 1 src
- Claude Code says 'Done' but it doesn't work, lecture explains · 1 src
Comments
via GitHub Discussions