AI Is Taking Over Accounting—Discover How Startups Are Making You Obsolete!

As artificial intelligence (AI) increasingly permeates various sectors, a new wave of startups focused on LLM observability is emerging to help organizations navigate their growing investments in this technology. These companies specialize in providing tools that allow businesses to monitor, debug, and optimize their AI deployments, crucial as spending on AI rapidly escalates.

According to research from Menlo Ventures, enterprise AI spending is projected to triple from $11.5 billion in 2024 to a staggering $37 billion in 2025. This rapid growth highlights the urgency for companies to scrutinize their AI expenditures and assess how these costs translate into tangible outputs.

Marco Argenti, Chief Information Officer at Goldman Sachs, emphasized the magnitude of these costs during a recent episode of Bloomberg’s Odd Lots podcast. He stated, “We’re going to have to accept that this is going to be a major item of cost in any organization, and it’s to be compared to the cost of people, not to the cost of TCP/IP packets or compute.” This statement underscores a significant shift in how businesses are approaching the budgeting and valuation of AI technologies.

Yet, many organizations currently face a lack of visibility regarding their AI expenditures. Derek Hernandez, an analyst at PitchBook, noted, “Most companies have limited visibility into where their AI spend is going, which models are delivering value, and where they’re burning tokens on low-impact tasks.” This gap in understanding is fueling a growing demand for observability tools, which are expected to gain traction as AI usage continues to proliferate.

This burgeoning market is reflected in venture capital investments. The total deal value for infrastructure SaaS and developer operations startups surged to $6.2 billion in capital across 133 deals since 2023, a 55% increase from $4 billion invested in 157 deals in 2024, according to PitchBook data. Notably, AI observability startups are leading the charge in this sector.

For instance, Braintrust, a startup providing solutions for tracking AI performance and token spending, secured an $80 million Series B funding round in February, backed by investors such as Iconiq, Andreessen Horowitz, and Greylock. Similarly, LangChain raised $125 million at a valuation of $1.25 billion last October, with support from prominent firms like Sequoia, Benchmark, and Amplify.

Moreover, OpenRouter, a startup designed to assist developers in tracking spending across multiple AI models, is reportedly in negotiations to raise $120 million at a valuation of $1.3 billion. This influx of capital into observability tools signals a critical shift in the AI landscape.

Despite a decrease in token prices—falling by 300x for commodity models and 12x for frontier models—overall spending on AI has risen. The reduction in token prices has facilitated more use cases, leading to increased complexity and resulting in higher bills for companies. OpenAI’s latest research revealed that the average reasoning token consumption per organization has surged 320x from 2024 to 2025.

The convergence of these factors—the rapid growth in AI spending, the demand for better observability tools, and the increase in token consumption—indicates a crucial moment for enterprises looking to maximize the value of their AI investments. As businesses adapt to this evolving landscape, it will be essential for them to engage with the right tools and strategies to ensure that they are not just throwing money at AI technologies, but instead are deriving meaningful insights and value from their expenditures.

In summary, as the AI market evolves, the emergence of observability startups represents a vital step in helping companies understand their AI spending and optimize their operations. The next few years will likely reveal how effectively organizations can leverage these tools to extract value from their investments in AI.