A‑I Tech News Today — Briefing

Hosts: Aurora with Isabelle
Date: 2026-07-06T20:26:15.000Z

Quick headlines — 4 fast stories

1) Google folds computer use into Gemini 3.5 Flash

Short summary: Google has integrated “computer use” directly into Gemini 3.5 Flash so agents no longer need a separate model.

Key points:

• Computer use is now a built-in tool alongside Search and Maps grounding.
• Agents built with the Gemini API can see, reason, and act across browsers, mobile, and desktops.
• Enterprise safeguards shipped with the feature:
  • Required user confirmation for sensitive actions.
  • Automatic task stopping when indirect prompt injection is detected.
  • Adversarial training to reduce risk.

2) Researchers document agentic ransomware called JadePuffer

Short summary: Sysdig reports the first documented case of agentic ransomware, dubbed JadePuffer, where an LLM orchestrated the attack.

Key points:

• Sysdig’s director of threat research, Michael Clark, says the attacker techniques weren’t new, but the A‑I’s coordination and automated ransom note lowered the skill floor.
• The agent:
  • Searched servers for A‑I API logins, cloud credentials, cryptocurrency wallets, and database keys.
  • Left human-readable commentary in code that aided attribution.

3) Utah pilots automated prescription refills with Doctronic — doctors push back

Short summary: A Utah regulatory sandbox lets residents request refills through an A‑I chatbot called Doctronic; clinicians currently review orders but a move to full automation is expected.

Key points:

• The pilot runs under a state A‑I regulatory sandbox that waives some rules.
• Oversight is by a five-member A‑I board that currently includes no doctors.
• The company expects to move from clinician review to fully automated refills.
• The state medical licensing board has asked for a halt, citing risks such as dangerous drug interactions (notably blood thinner refills).
• The FDA is currently taking a hands-off stance.

4) Big tech softens on the “jobs wipeout” narrative

Short summary: After dire predictions about mass job losses, several industry leaders have moderated their rhetoric.

Key points:

• Sam Altman (OpenAI) said the sector underestimated how central people would remain.
• Dario Amodei (Anthropic) has emphasized more positive scenarios while still warning of risks.
• An EY‑Parthenon survey found CEOs expecting large A‑I-driven head-count cuts dropped from ~46% (January last year) to ~20% (May this year).

Thanks for listening.
Stay curious, stay skeptical.
Subscribe to A‑I Tech News Today and we’ll bring the headlines to your ears.