Tag
This paper evaluates the reliability of Gemini models as audio judges for scoring full-duplex voice agent conversations, finding that Gemini 2.5 Flash shows strong agreement with human raters on most dimensions, though model swaps require re-validation.
The paper presents Libra, a self-evolving framework that introduces mutable catalogs into repositories to improve code localization for LLM agents, achieving logarithmic improvements and zero-shot transfer across different models and problem sets.
ServiceNow's customer chief calls 'tokenmaxxing' an AI hype cycle, criticizing the focus on token volume as a metric, while Salesforce shifts to $2 per resolved issue pricing, emphasizing outcome-based measurement.
Salesforce employees are confused why the company is promoting Anthropic's Claude Tag, a rival AI teammate inside Slack, as it competes directly with Salesforce's own Slackbot and Agentforce platforms, despite deep financial ties between the two companies.
Salesforce announced a $2 per resolved AI agent issue pricing model, defining 'resolved' as completion without human escalation or customer abandonment. This outcome-based pricing contrasts with token metering and may set a precedent for enterprise AI costs.
Gergely Orosz observes that Salesforce heavily promotes Agentforce for AI, yet most tech companies ignore it and instead use Slack as their AI connective tissue, suggesting Salesforce won AI infrastructure but failed to notice Slack's importance.
Box CEO Aaron Levie argues that AI agents will use software 100X more than people, requiring guardrails, authoritative data sources, logging, and collaboration features; platforms enabling headless interactions will be best positioned.
Katalyst is an AI agent designed to automate Salesforce pipeline management.
Salesforce acquires AI customer service platform Fin for $3.6 billion to enhance its Agentforce platform with advanced AI agent capabilities.
Salesforce has agreed to acquire Fin (formerly Intercom), a customer agent platform, for $3.6 billion to accelerate its autonomous agent capabilities and expand agentic enterprise offerings.
Analysis of how enterprise software vendors like ServiceNow, Microsoft, and Salesforce are racing to build AI governance layers to avoid becoming irrelevant middlemen, treating governance as a strategic moat rather than mere compliance.
Salesforce deployed 20,000 enterprise AI agents, revealing that the majority of effort comes after launch, not before. John Kucera, CPO of Agentforce, shares lessons on what separates successful agents from those that stall.
Based on a16z's analysis of Salesforce's headless products, this article explores the trend of enterprise software moats shifting from user interfaces to underlying data models, permission systems, and workflow logic in the AI Agent era, and analyzes the difficulty differences in migrating CRM, ATS, ERP, and other systems.
Discussion of Salesforce's CRMArena-Pro benchmark showing agent success drops from 58% on single-turn to 35% on multi-turn tasks, plus practical advice for splitting agent workflows into narrow stages to reduce error compounding.
Bloomberg investigation reveals gap between Salesforce's Agentforce marketing and actual enterprise deployment, highlighting broader industry challenge of AI demos outpacing operational reality.
Salesforce CEO Marc Benioff revealed the company will spend $300M on Anthropic tokens this year, has hired zero software engineers since January 2025, and cut support staff from 9,000 to 5,000 using AI agents. Agentforce reached $800M ARR, up 169% YoY, signaling a structural shift from payroll to token spend.
Salesforce announced headless, opening API for AI agents to directly read and write databases, bypassing UI. This is seen as a signal that the traditional moat of the SaaS industry is collapsing.
A new paper by Microsoft Research and Salesforce reveals that LLM performance drops significantly in multi-turn conversations due to a 'Lost in Conversation' phenomenon, challenging the reliability of current single-turn benchmarks.
Salesforce CEO Marc Benioff announced a plan to hire 1,000 new graduates and interns through a new Builder Program focused on the company's AI agent system, Agentforce.
Salesforce integrates OpenAI's models into its Einstein AI Platform with enterprise-grade security features, offering generative AI capabilities for Sales, Service, and Commerce Cloud with the new Einstein Trust Layer to ensure data privacy and compliance.