Generative AI Enterprise
Posts
Will Agentic AI eliminate SaaS?

Will Agentic AI eliminate SaaS?

Plus, shifting your mindset, AI agent evaluation, and more.

Lewis Walker
January 26, 2025

WELCOME, EXECUTIVES AND PROFESSIONALS.

One of the most intriguing questions being debated today is whether AI will spell the end of traditional software as a service (SaaS). Bain & Company offers insights on what past technology disruptions can teach us about the answer.

Since the previous edition, we've reviewed hundreds of the latest insights on best practices, case studies, and innovation. Here’s the top 1%...

In today’s edition:

Will Agentic AI eliminate or enhance SaaS?
Sequoia on adopting the stochastic mindset.
OpenAI's web agent brings enhanced autonomy.
Galilieo shares AI agent evaluation best practices.
Transformation and technology in the news.
Career opportunities & events.

Read time: 4 minutes.

PLATFORM STRATEGY

Will Agentic AI eliminate or enhance SaaS?

Image source: Bain & Company

Brief: Bain & Company Partner Chuck Whitten explores whether agentic AI will eliminate or enhance SaaS. History, he argues, shows technological revolutions tend to expand ecosystems rather than replacing them outright.

Breakdown:

Leaders like Satya Nadella suggest agentic AI could upend the SaaS model entirely. Others believe AI will simply enhance SaaS.
Technology revolutions are rarely binary: client/server computing didn’t completely eliminate mainframes, neither did cloud with on-prem systems. Instead, they transformed and coexisted.
If AI replaces the “business logic” layer of SaaS, agentic tool builders still face a steep challenge: mastering the nuanced use cases SaaS refined for years.
To succeed, they need to establish trust and transparency with customers. Businesses won’t adopt tools they can’t control or verify, and adapting AI to highly specific workflows is nontrivial.
SaaS vendors may embed AI deeply, creating hybrid solutions that merge AI agents’ intelligence with specialized SaaS strengths.
Many incumbents’ technical roadmaps aim for this, but it demands a willingness to disrupt their own businesses from within.

Why it’s important: SaaS disruption could redefine enterprise tech, partnerships, and ultimately those who lead the market. Bain’s Whitten predicts convergence, not collapse, where AI and SaaS both complement and compete. The lesson from history: Transitions expand ecosystems rather than replace them outright.

WAYS OF WORKING

Sequoia on adopting the stochastic mindset

Image source: Sequoia

Brief: Sequoia, a leading Silicon Valley VC, suggests knowledge workers should adopt a "stochastic mindset" leveraging AI to accelerate information processing while critically interpreting less predictable model outputs.

Breakdown:

AI is reshaping how we experience and think about work.
Knowledge workers typically relied on technology with predictable outputs. AI brings probabilistic or "stochastic" outputs, where results are not guaranteed.
This shift moves us from step-by-step workflows to iterative, experimental approaches, like delegating tasks where you gain speed but lose control over the exact outcome.
With AI, we can process more information faster, but it also means dealing with uncertainty in outputs, requiring stronger critical thinking skills.
As AI reasoning increases, outputs may become even more unpredictable, as OpenAI co-founder Ilya Sutskever recently noted.
Knowledge worker adaptability is increasingly key. Small, agile teams can thrive by experimenting and iterating quickly with AI.

Why it’s important: The stochastic mindset isn’t a new way of thinking, but the rise of AI makes it much more important. Less line workers and strategic managers. Less programming and more teaching. The stochastic mindset will help make us all more comfortable at a higher level of abstraction.

INNOVATION INSIGHT

OpenAI's web agent brings enhanced autonomy

Image source: OpenAI

Brief: OpenAI launched Operator, an AI agent that independently navigates web browsers to complete tasks, marking a major step forward in the company’s pursuit of autonomous agentic systems.

Breakdown:

Operator uses OpenAI’s new Computer-Using Agent model that combines vision capabilities with advanced reasoning to “see” (via screenshots) and interact (using mouse and keyboard) with a browser.
OpenAI demoed the technology live, showcasing tasks like booking reservations, ordering groceries, and buying event tickets.
If Operator encounters challenges, it attempts to self-correct using reasoning. If stuck, it hands control back to the user.
Built-in safety features include user approval for purchases, automated threat detection, and "takeover mode" for sensitive info like passwords.
The research preview is currently limited to U.S. Pro users, with plans to expand to Plus, Team, and Enterprise after further testing.
Operator is still learning and evolving, and can make mistakes. For instance, it can face challenges managing calendars.

Why it’s important: Agentic systems are emerging more frequently, with examples like Anthropic’s Computer Use, Self-Operating Computer Framework, and Browser Use. OpenAI’s move marks a major step forward, with the potential to help enterprises automate increasingly complex tasks in the near future.

BEST PRACTICE INSIGHT & CASE STUDIES

Galileo shares AI agent evaluation best practices

Image source: Galileo

Brief: Galileo, a company that specializes in AI evaluation, released a 93-page guide on mastering AI agents. It covers agent capabilities, real-world use cases, and frameworks, with a strong focus on performance evaluation.

Breakdown:

Chapter 1 introduces AI agents, their ideal uses, and scenarios where they can be excessive. It includes real-world cases from Salesforce and Oracle Health.
Chapter 2 details frameworks: LangGraph, Autogen, and CrewAI, providing selection criteria and case studies of companies using each.
Chapter 3 explores how to evaluate an AI agent through a step-by-step example using a finance research agent.
Chapter 4 covers measuring agent performance across systems, task completion, quality control, and tool interaction, with five detailed use cases.
Chapter 5 addresses why many AI agents fail and provides practical solutions for successful AI deployment.

Why it’s important: As AI agents become more prevalent, ensuring they work correctly and safely is key. This is where evaluation comes in. Galileo’s previous guide focused on "Mastering RAG," building enterprise-grade systems. Now, they’ve taken it further with agents using LLMs to complete broader, more complex tasks.

Accenture shared its journey transforming its marketing function with AI agents, reducing the average number of campaign steps from 135 to 85 and speeding up time-to-market by 25-35%.

McKinsey announced a strategic alliance with C3 AI, combining the expertise of QuantumBlack and McKinsey’s AI practice with C3 AI's software to accelerate enterprise AI transformations.

LinkedIn published a case study on enhancing LLM training GPU efficiency, it focuses on an open-source library that can improve training throughput by 20% and reduce memory usage by 60% for models like Llama, Gemma, and Qwen.

The Allen Institute released The Generative AI Ethics Playbook a 63-page guide on identifying and mitigating risks in the design, development, and deployment of datasets and models.

Google shared a 70-page quick-start handbook on effective prompting, covering prompts for executive, marketing, customer service, and other roles.

Google DeepMind unveiled Gemini 2.0 Flash Thinking, a free experimental AI model that leads in mathematics, scientific reasoning, and multimodal benchmarks, ranked No. 1 on LM Arena’s leaderboard.

Microsoft revised its exclusive cloud deal with OpenAI, retaining first-refusal rights for computing capacity while allowing OpenAI to seek additional infrastructure partnerships.

Anthropic launched Citations via its API and Google Cloud’s Vertex AI, offering references to the sentences and passages used in responses, enhancing verifiability.

Google is reportedly investing an additional $1B in Anthropic, raising its total stake to $3B. A separate funding deal is also in progress, set to increase the company's valuation to approximately $60B.

Center for Security and Emerging Technology released a 29-page report on China's approach to pursuing artificial general intelligence (AGI), highlighting critiques of a singular LLM strategy and advocating for a more diversified approach.