Soon, there will be job websites with posts that only AI agents can apply for. 2025 will be the year AI colleagues enter our lives. But which company will provide these AI workers? To find out, let’s look at what companies need: access to users and access to business data. Which company will supply these AI colleagues? Let’s break down what companies need to become the main supplier of AI agents: access to users and access to enterprise data.
Microsoft’s Pole Position — Access To Users
Who offers all our workforce tools? Microsoft. Microsoft. Microsoft. It is in pole position for access to users. Most people already use Microsoft products — whether they like it or not — and AI is becoming deeply integrated into these tools, with Copilots appearing everywhere.
This market access is a big advantage. In 2023 and 2024, many startups launched impressive AI services, only to be quickly overshadowed by giants like Microsoft and Google, which have direct access to customers. Take, for example, Jasper.ai, a once-celebrated AI tool for writing text. As I pointed out in a LinkedIn post, similar features are now built directly into Google and Microsoft products, making it increasingly difficult for smaller players to compete.
OpenAI’s Desktop APP
This fate might also await OpenAI if they do not create their own tool or hook soon. The new desktop app might be that kind of hook. This app can now read code directly from developer-focused tools like VS Code, Xcode, TextEdit, Terminal, and iTerm2. This means developers and writers no longer need to copy and paste their code into ChatGPT — a common hassle for many of us. It’s a really useful tool and a smart move to integrate more deeply into the desktop user workflow. We’ll see whether this can chip away at some of Microsoft’s dominance.
The Power Of Data Access
AI needs data to be effective. Let’s say you’re looking for answers about a company’s internal processes or insights from documents. General tools like OpenAI’s ChatGPT won’t cut it. What we need are tools that can read and summarize company documents, tailored specifically for enterprise use. As I’ve said before, 2025 will be the year of search — especially enterprise search. Transformer models, such as those offered by OpenAI and others, could provide this ability but will require internal data.
Who has access to this kind of data? Microsoft, Amazon, and Google are major players here. They store data in platforms like SharePoint. But they’re not alone. Salesforce, for example, also holds an enormous trove of valuable data — customer interactions, discussions, process documents, marketing strategies, and more. Does Salesforce want AI agents to help unlock this potential? Absolutely. Are they currently the underdog? Yes.
Therefore, it’s no surprise that Salesforce CEO Marc Benioff recently took a jab at Microsoft. He called its AI assistant, Copilot, “disappointing,” saying, “It just doesn’t work, and it doesn’t deliver any level of accuracy.” He even dubbed it “Clippy 2.0” — the funniest insult I’ve heard in a while — before introducing Salesforce’s own AI solution, Agent Forces.
And OpenAI? How Can They Compete?
OpenAI doesn’t have the same level of data access or consumer reach as Microsoft, nor does it have Salesforce’s treasure trove of business data. So, what’s its angle? OpenAI claims to have the smartest tools on the block — and it probably does, although I personally find Anthropic’s Claude 3.5 better than OpenAI’s GPT-4 at the moment.
OpenAI is betting on its ability to outperform everyone else with superior technology. The current frontier of technology? Reasoning — essentially, models that “think longer,” as I mentioned in my comments about Sam Altman’s recent post on AGI.
Chatting With Large Language Models Costs Money
“Thinking longer,” however, means more CPU usage and higher costs. OpenAI recently raised $6.6 billion in a Series E funding round — a much-needed boost to sustain its operations. While Salesforce’s Agentforce generates solid revenue from its customers and Microsoft enjoys a massive financial war chest, OpenAI is still in the early stages of convincing businesses and users to pay enough to offset the steep costs of developing cutting-edge AI.
Their $200-per-month premium tier, which includes the expanded version of O1, is a step in this direction. But is it worth the price? Reports suggest that a well-designed prompt — meaning a human writes the instruction — can, so far, outpace OpenAI’s o1.
That will change, but it’s also clear that OpenAI won’t be the only player offering advanced reasoning. o1 could soon be outpaced by open-source alternatives, as we’ve seen before with Meta’s Llama.
Speaking of Meta, one of my 2025 predictions is that they will attempt to monetize AI models next year. Ultimately, they too will face the challenge of justifying enormous costs without securing a steady and reliable revenue stream.
AI Agents Will Be in the Workforce
In 2025, we’ll see more AI agents entering the workforce, transforming workflows by simplifying, enhancing, and automating tasks across industries. These won’t be all-encompassing AGI models but smaller, specialized models designed for dedicated workflows. AI will scale and improve processes one step at a time, combining traditional AI, context retrieval (RAG), and robust user design to address challenges like security, hallucinations, and user control.
Amazon, Salesforce, Google, and Microsoft will all try to become the underlying technology platform for these AI agents. For OpenAI and Sam Altman, the key strategic question will be how to price OpenAI’s base models for Microsoft or Salesforce if they end up competing directly with them.
