A forced API test across 479 U.S. election races produced sharply different candidate picks from Grok and GPT. What does that reveal about AI—and what does the test leave unanswered?
Oscar Gallo and Matt Wozniak unpack the test's limitations and why an API response can differ from the consumer chatbot experience. The conversation moves through OpenAI coding agents interfering with safety monitoring, the Decisions API announcement versus its documentation, Anthropic's Agent Skills donation, and AWS desktop sessions for AI agents.
In No Jargon Required, we explain what an AI Agent Skill is and how consumer chatbots differ from APIs. Then Unpopular Opinions opens a debate about AI companies, public trust, and when safety rules become editorial control. These are the hosts' arguments and opinions, with room for disagreement.
Two concepts behind this week's stories.
What an AI Agent Skill is, and why it matters that Anthropic donated the format to the Agentic AI Foundation.
How the consumer chatbot differs from the API, and why an API test can return different answers than the app people actually use.
- Convovo candidate test
- TechCrunch coverage
- OpenAI monitoring incident report
- BleepingComputer coverage
- Decisions API announcement
- OpenAI changelog
- Decisions API guide
- Anthropic Agent Skills announcement
- Linux Foundation announcement
- AWS WorkSpaces announcement
- AWS agent-access guide
- Agent Skills guide
- OpenAI text-generation guide
- xAI Chat Completions guide