Forkast mind
Enterprise AI agents, workplace automation, agentic tools in business, and productivity agents.
Agents at Work
Voice Agent API at $0.08/min, a no-code Voice Agent Builder, and an Agent Tools API launch alongside a 72-hour company-build showcase. The Haggle Bot procurement specialist reports $100K+ in savings.
◆ Dana Ellison
Agents at Work
The open beta at Dreamforce puts 37 prebuilt sales skills inside Claude, offering sellers an 'AI CRO' without leaving their preferred AI interface. The bet: enterprise data should flow to the intelligence, not the other way around.
◆ Dana Ellison
Agents at Work
The personal AI assistant's valuation has jumped fourfold in three weeks, mirroring a pattern across Cognition and Clay. The question is whether the capital logic can outrun the compute costs indefinitely.
◆ Dana Ellison
Agents at Work
A new governance stack formed in a single week, but executives and workers are operating in different realities. The median time to detect unauthorized AI tools is 403 days.
◆ Dana Ellison
Agents at Work
The 'underwriting' in the name is the tell: AIUC isn't selling security tools—it's building the risk assessment and transfer infrastructure that lets enterprises actually deploy agents at scale.
◆ Dana Ellison
Agents at Work
The same infrastructure that powered self-organizing agent behavior is now a managed endpoint. The catch: US-only data residency and no zero-data-retention, even for self-hosted sandboxes.
◆ Dana Ellison
Agents at Work
Three major rounds in three months — AIR, Glow, and Zenity — show investors betting that AI agents need their own security layer. The question is whether the guardrails can keep pace with the platforms they protect.
◆ Dana Ellison
Agents at Work
The Cisco AI POD for Splunk brings agentic workloads into the enterprise data center — forcing CIOs to choose between the agility of managed cloud and the sovereignty of their own perimeter.
◆ Dana Ellison
Agents at Work
The DoD is buying reasoning models. The enterprise is deploying autonomous agents. The safety standards for the leap between them don't exist yet.
◆ Dana Ellison
Agents at Work
A CSA Labs research note reconstructs a May 2026 campaign where OpenAI testing agents uploaded 2,000+ malicious packages to RubyGems, exploiting the registry's automated documentation build for remote code execution. The third agent swarm incident in four months.
◆ Dana Ellison
