MCP for agent-to-agent comms may be the riskiest protocol you've never heard of
The adoption of AI agents in millions of organizations is creating new opportunities for attackers to make them take malicious actions, such as exfiltrating…
According to Ars Technica, the adoption of AI agents in millions of organizations is creating new opportunities for attackers to make them take malicious actions, such as exfiltrating database contents and sensitive business and personal information.
In the past five months, Google and four other organizations—with little in common except for their use of AI agents—have acknowledged vulnerabilities that exploit one agent inside a targeted network to spread harmful instructions to other internal agents.
This technology story centres on the United States. Names and organisations mentioned in the coverage include MCP, AI and Google.
Why it matters
Technology stories like this can affect how people work, communicate and protect their data.
Key details
- Location
- United States · Americas
- Reported by
- Ars Technica
- First reported
- Key names
- MCP, AI, Google, LLM
- Section
- Technology
Sources
This is an independent summary. Read the full, original reporting from the publisher:
Read the full story at Ars TechnicaSpotted an error? Request a correction · How this summary was produced
Related Stories
Walmart says AI training is boosting confidence with the technology among its 2.1 million associates
Walmart says training from OpenAI and Google is helping its 2.1 million associates feel more confident using AI and digital tools.
Canva’s first AI image model plans designs before generating them
Canva is diving even deeper into AI with the launch of an image generation model that competes directly with those from OpenAI, Microsoft, and Google.
Trump hints at ban on US firms building AI data centres abroad
President Donald Trump warned U.S. tech companies they could face a huge tax and possibly a ban if they build AI data centres overseas.
Global stock markets slump as Anthropic IPO fails to stem AI sell-off
Asian markets on track for their worst week since 2020, with European and US stocks also way down Business live – latest updates Global stock…
Nations agree to ‘elaborate’ on rules for AI-run ‘killer robots’ by 2028
Members of the UN's Convention on Certain Conventional Weapons fail to agree on swift adoption of legally binding rules.
FBI probes possible breach of national crime database
The FBI is said to be investigating an apparent hack of the National Crime Information Center, the BBC reported.
Al Jazeera explains: The US-China AI race is entering a new phase
Al Jazeera’s Matthew Torres unpacks the latest from the world of artificial intelligence.
Trump urges Nvidia to reinstate Jensen Huang and demand apology
Al Jazeera English reported that former US president Donald Trump called Nvidia, asking the company to give former CEO Jensen Huang a second chance.



