7 Business Lessons For AI

This is an approach many are eager to get right.

They encounter several hurdles—the technology is still in its infancy, there’s a lack of clear roadmaps for businesses, and widespread skepticism about using large language models (LLMs) to automate core operations.

So what should leaders do?

A Leader’s Insight

I recently gained valuable perspective from a paper by Lidiane Jones, former head of Slack and CEO of Bumble, a leading dating platform.

Jones outlines key considerations that executives are weighing when integrating AI into their organizations.

Data Integration and Oversight

She notes that while data processes like ETL (extract, transform, load) and ELT (extract, load, transform) existed long before AI, data silos remain a persistent issue.

Her proposed solution? An “omnichannel data strategy.” According to Jones, this approach “ensures data privacy and security, simplifies access for business applications, enables real-time functionality, and seamlessly integrates with everyday tools.”

Adhering to Financial Compliance Standards

Jones emphasizes the importance of compliance, particularly with financial data.

“Every organization handles sensitive financial information governed by audits, regulations, and compliance requirements that must be rigorously safeguarded,” she states. “In larger enterprises, this data typically resides in ERP systems. CEOs, CFOs, COOs, and CROs all depend on timely insights from these systems to assess performance against goals, monitor spending versus budgets, or evaluate how shifts in workforce investment might impact overall costs and business momentum.”

Empowering Decisions with Business Intelligence

On the topic of business intelligence, Jones shares a compelling example:

“Picture a Sales Executive who has nurtured a multi-year, high-trust relationship with one of the company’s top clients—and then decides to leave for a new opportunity,” she writes. “While there may be records and notes about the client, much of the deep, contextual knowledge walks out the door with her. Organizational wisdom is shaped by people, culture, and processes—and often resides in individuals.”

She highlights how collaboration platforms can help preserve this knowledge.

“Tools like Slack, Google Workspace, and Microsoft Teams contain vast amounts of embedded human insight that are rarely leveraged,” she observes. “Unstructured data from these sources is incredibly valuable for training LLMs, unlocking new possibilities—such as identifying what aspects of the sales relationship this key client valued most.”

She acknowledges the challenge: “Expanding data strategy to this scale might seem overwhelming,” but suggests that collaboration with vendors and partners can ease the transition.

“Phasing and prioritizing the integration of your data into a unified system is essential to making steady progress and delivering tangible business value,” she advises.

AI Agents Take Action

Jones underscores the transformative potential of AI agents. Unlike traditional systems that process and present data, these agents can actively brainstorm and reconfigure workflows.

“In many cases, agents can self-optimize processes by discovering more efficient ways to complete tasks,” she explains.

A Real-World Journey

Referencing ChatGPT’s rapid ascent, Jones reflects on her time at Slack—a platform built for business communication—and how early partnerships shaped its AI evolution.

“When I was leading Slack, it was energizing to work with OpenAI, Cohere, and Anthropic, leveraging their LLMs to tackle some of our customers’ toughest productivity challenges,” she recalls.

These included: “locating a forgotten conversation in a sea of channels, summarizing and prioritizing message overload, improving search functionality for faster information retrieval, and more.”

The team also developed new tools.

“We launched Slack Canvas with templates to help users generate content using company data, and began capturing meeting notes and action items from Huddles—just the beginning,” she says. “LLMs allowed us to address real customer pain points in intuitive, meaningful ways, all while preserving the essence of the Slack experience.”

Jones describes such capabilities as the “table stakes” of today’s corporate landscape, paving the way toward what she calls “co-intelligence”—a future driven by workflow automation, agentic AI, multi-agent systems, and innovative interfaces.

Managing Our Digital Workforce

Jones offers a powerful metaphor for handling AI agents:

“Viewing autonomous agents as ‘digital workers’ can help frame familiar management questions: How does the organization assess the quality of their output? What systems are they authorized to access? During an audit, how do we trace the actions taken by a digital worker? And if an agent behaves maliciously, how do we shut it down?”

On autonomy, she clarifies that even fully independent agents need clear parameters.

“Like humans, autonomous agents require a defined job scope or set of instructions to operate effectively,” she notes. “This is a reality leaders must prepare for—soon.”

Four Leadership Principles for the AI Era

Jones concludes with actionable guidance for leaders navigating AI adoption:

1. Be hands-on: Stay actively involved in AI initiatives as a leader
2. Collaborate strategically: Partner with vendors and external experts
3. Cultivate an AI-first mindset: Launch AI-native projects to drive innovation
4. Focus on value: Identify and prioritize use cases that deliver real business impact

This framework offers practical wisdom for anyone planning a major strategic shift in the AI era. Some of the most impactful leadership insights come not from textbooks, but from real-world experiences—shared through TED talks, industry events, and firsthand accounts like Jones’s.

The above is the detailed content of 7 Business Lessons For AI. For more information, please follow other related articles on the PHP Chinese website!