StaffAI ads

What’s the difference between an AI employee and an AI agent? Simple answer: one has a seat on the org chart. The other has a seat in your tech stack. The longer answer: join us 10/29 when Michael Adams from Pop Sells will share how he has onboarded a team of AI employees that have allowed his sales department to double their conversion rates without additional headcount. Join this session to learn: • Which roles can be fully automated with AI employees, and where human employees remain key contributors • What AI employees cost vs. their human counterparts • How to keep proper governance when AI employees talk directly to customers This is our 22nd AI Roundtable at the Alliance of Chief Executives, and each one brings 70+ CxOs who learn from each other’s real AI initiatives. If you lead a non-tech business with 20–500 employees, this session is designed for you. RSVP link in the comments. See you next week!

AI is moving so quickly that it can be a full-time job just to keep up-to-date. For busy executives, the key is to find signal in the noise. That is what we provide at the Alliance of Chief Executives for mid-market enterprise leaders 3 times a year in our State of AI Adoption Roundtable. Join us next Thursday, September 24th where we’ll be covering: • AI adoption in the mid-market by industry, departments and use cases • Ranking of AI model providers by capability, cost and use case • Our favorite demos of AI for the mid-market If you are a non-tech business leader of a company with 20-500 employees this roundtable is designed for you. Seema Alexander and I will share industry benchmarks as well as our collective learnings having worked with dozens of mid-sized companies. Join us for the first 30 minutes to see the full slides below, and stay for the hour to get the real learnings in breakout sessions where you’ll hear candid adoption stories from your peers. RSVP in the comments below, and see you next week!

How big is the AI bubble? The Economist presents an insightful approach to measuring the bubble based on the gap between data center infrastructure capex being invested by the "hyperscalers" vs revenues being collected by the "AI model providers" who are using that infrastructure. In their assessment, the revenue gap is roughly $2.3trn/year based on the $150B of identifiable AI revenue today against the ~$2.5trn/year required to cover current capex plans i.e. spend needs to increase from ~$150B to ~$2.5trn, a ~16× jump, in order to deflate the bubble. As the Economist notes, the current buyers of AI are almost entirely businesses (rather than consumers), and of enterprise buyers its currently mostly other tech companies. The key question is what will get non-tech companies to adopt AI in order to reach meaningful levels of AI spend? In this article, we'll explore three of the leading AI delivery models in order to evaluate how likely AI spend can catch up with the AI infrastructure investment boom.

Last week Andreessen Horowitz came out with excellent research on the state of Computer Use AI. Here are the core findings: • The hourly rate to run a Computer Use model is now less than an offshore worker • Computer Use AI beats humans in accuracy scores for many desktop tasks • Enterprises are running millions of automated Computer Use tasks a month in production So what's next? In this article we explore how enterprises are adopting Computer Use in the context of "AI employees”.

90% of revenue comes from repeat customers for the typical B2B enterprise. That's true for Salesforce, and it's true for your local HVAC distributor. The difference is that SaaS automates nearly 100% of this repeat revenue: the contract renews, software charges the price, the product provisions itself. No humans required. For everyone else (distributors, brokers, equipment dealers) a human rep sits inside nearly every repeat transaction: the order needs configuring, the price needs negotiating, exceptions need handling. Rules-based software hasn't worked for these types of "high-touch" interactions, so these businesses grow one salesperson at a time, while SaaS grows on code. That gap is why software trades at ~10x the revenue multiple of high-touch businesses. AI employees that use "reasoning models" change the equation. In our latest post, we profile how these AI employees are now able to automate high-touch order configurations, pricing negotiations and contract facilitation at scale. Read more in the link below. https://lnkd.in/gZHMRHk2

A sales rep at a traditional mid-sized enterprise might spend as much as 2–3 hours per day finding leads to work. Many of these leads sit behind gated portals that don’t have an API (contractor databases, procurement portals, Facebook Marketplace, etc). The only way to collect these leads is by manually clicking through the website because web scrapers get blocked. This is where AI thrives. A StaffAI “Inside Sales Rep” does the work in the exact same way. It does this through “computer use” and clicks around these web portals just like a person would. First, it watches how a human sales rep does their task (where they click, what search filters they use, etc). It then learns how to replicate the same mouse clicks and keyboard strokes. All the human rep needs to do to train the AI is perform their work as normal (no programming required, and no GUI builder to learn). If you think the video is boring, that’s the point! Most of this work is necessary but dull. Now AI can find great leads on its own so human reps can have more time for “higher touch” phone and in-person meetings. Expect to see AI “embodying” the human workforce in the coming months; but no need to wait, get in touch if you’d like to see how AI can operate your desktop work today.