Research
A small number of samples can poison LLMs \ Anthropic
Anthropic research on data poisoning attacks shows a few hundred malicious documents can "backdoor" large language models of any size.
ResearchProject Pilot: Can AI models fly drones? \ Anthropicanthropic.com
ResearchProject Fetch: Can Claude train a robot dog? \ Anthropicanthropic.com
ResearchEmergent introspective awareness in LLMs \ Anthropicanthropic.com
ResearchClaude Opus 4 and 4.1 can now end a rare subset of conversations \ Anthropicanthropic.com
ResearchLLM-discovered 0 days \ Anthropicanthropic.com
ResearchMoral self-correction in large language models \ Anthropicanthropic.com
BlogsBreaking down common documentation mistakesMintlify
LearnCalling Your Video Game With Your Phone: Part 1Deepgram
BlogsThe Best (and Worst?) of Voice AI TechnologyDeepgram
BlogsHow to Conduct an AI Agent Security Audit [+ Checklist]Zapier
BlogsHow to implement AI training for employeesZapier
Blogs7 ways to use Zapier's Zendesk integrationZapier
BlogsHow to connect Google Sheets to NotionZapier
NewsHow Argo CD and OpenShift Enable GitOps for DevelopersThenewstack
NewsWhen CI Meets CD: Deliver With ConfidenceThenewstack
NewsOData or GraphQL? The Best Tech for Developing an API Is Neither or Both!Thenewstack
BlogsPulumi ESC and External Secrets Operator: The Perfect Solution for TodayPulumi
BlogsIntroducing New Slimmer Docker ImagesPulumi
ResourcesThe Go Memory Model - The Go Programming LanguageGo
BlogsGo.dev: a new hub for Go developers - The Go Programming LanguageGo
