What is at stake, in 400 words ?
AI agents use artificial intelligence to achieve goals by making decisions, taking actions, and performing tasks
The more narrowly defined and targeted, the more effective the agents will be
Business software applications, 'Software as a Service' (SaaS) - either customer fronting (CRMs) or back-office processing (ERPs) - rely on arrays of specific functionalities, interacting with one another to automate tasks
This is also a critical, and potent, feature of AI agents...
Software providers have essentially remained in their lane of expertise, with limited overlap between front- and back-office services
However, the temptation to broaden the scope from front- to back-office (and the reverse) always existed because business applications disentangle and analyze the data of the client company
And data is the defining source shared by CRMs and ERPs
The key question, going forward, is how tightly legacy SaaS systems can hold onto client data
If – and only if – clients can take back control over their own data architecture, the SaaS business model will be at risk over time
AI agents have some way to go to be securely aligned on their assigned goals, secure and compliant
Fast paced progress of the AI agents, and their ability to define their own data structure with company data, might be truly transformative – and questionable
By proceeding stepwise with workflows around existing SaaS systems, companies would start gaining direct and pointed control with AI agents – a real possibility and an attractive proposition
The speed with which AI agents will invade SaaS applications is moot…time will tell
The emergence of AI will impact SaaS business models, forcing service providers and clients to focus on undisputable value-added - not a bad thing...
SaaS - Software as a Service - defines itself as a service provider, intermediating a company's data pool, providing readability, structure and interpretation
In the fast-evolving AI landscape, lean approaches appear to move away from massive scaling up of computing power and data availability, the defining features of the immensely costly large language models (LLMs)
Disintermediation, boosted by lean AI data exploration, will remain tentative as trust in the reliability of direct processes builds up across companies
This assumption, if true, may only gain a foothold at first, as SaaS providers actively protect their legacies....
Our theme "Will AI agents replace software apps?" highlights how the software firms respond to the challenge
By tracking the performance of Customer Relation Management providers, such as Salesforce
***
A side note
From Business Insider, September 15:
- Walmart is using AI to solve vast and complicated physical challenges.
- The company's mix of digital and physical operations is making it an unexpected AI powerhouse.
- Walmart has created new AI "super agents" to help users manage the proliferation of tools.
"
"Analysts have characterized the recent strength in the stock market as an AI rally, but flying under the Magnificent Seven's radar is Walmart — a company so vast that it literally has its own weatherman.
And as it turns out, the retail juggernaut's scale and reach are proving to be tremendous assets in the AI race......"