Latest News
CRED uses AI with Cleo, Thea and Stark to build concierge‑level service OpenAI goes for-profit; Cursor 2.0 shifts to in-house AI; Anthropic adds Claude Google Gemini's Deep Research can access Gmail, Drive, and Chat for AI reports AI data center surge inflates Silicon Valley spending, earnings and accounting AI Capacity Crunch: 90% of Code Now Agent‑Generated, Latency and Cost Rise Apple to Deploy Custom Google Gemini Model for Apple Intelligence German Commons opens pipeline to free AI datasets from copyright limbo AgiBot, Chinese startup, pays workers to teleoperate robots for US data demand CRED uses AI with Cleo, Thea and Stark to build concierge‑level service OpenAI goes for-profit; Cursor 2.0 shifts to in-house AI; Anthropic adds Claude Google Gemini's Deep Research can access Gmail, Drive, and Chat for AI reports AI data center surge inflates Silicon Valley spending, earnings and accounting AI Capacity Crunch: 90% of Code Now Agent‑Generated, Latency and Cost Rise Apple to Deploy Custom Google Gemini Model for Apple Intelligence German Commons opens pipeline to free AI datasets from copyright limbo AgiBot, Chinese startup, pays workers to teleoperate robots for US data demand
Our content generation service is experiencing issues. A human-curated summary is being prepared.
Business & Startups

AI Capacity Crunch: 90% of Code Now Agent‑Generated, Latency and Cost Rise

By
• 2 min read

Why does the AI capacity crunch matter now? Developers are watching latency creep up while cloud bills swell, and the headline numbers are stark: 90% of code is being churned out by autonomous agents in certain workflows. While the tech is impressive, the cost side‑effect is less celebrated.

Companies that once relied on human coders are feeling the pinch of longer response times and higher compute spend, a trend that some analysts trace to agents hitting their operational limits. But here's the reality: as agents mature, the conversation in top labs is shifting. Reinforcement learning, once a niche research topic, has become the hot‑button issue among data scientists at OpenAI, Anthropic and Gemini.

They’re treating it as a potential lever to stretch capacity further, even as pricing models edge toward surge‑pricing thresholds. The tension between productivity gains and mounting expenses sets the stage for the observation that follows.

It's now estimated that in some cases, 90% of software is generated by coding agents. Now that agents have essentially come of age, Bercovici noted, reinforcement learning is the new conversation among data scientists at some of the leading labs, like OpenAI, Anthropic, and Gemini, who view it as a critical path forward in AI innovation.. It blends many of the elements of training and inference into one unified workflow," Bercovici said.

"It's the latest and greatest scaling law to this mythical milestone we're all trying to reach called AGI -- artificial general intelligence," he added. "What's fascinating to me is that you have to apply all the best practices of how you train models, plus all the best practices of how you infer models, to be able to iterate these thousands of reinforcement learning loops and advance the whole field." The path to AI profitability There's no one answer when it comes to building an infrastructure foundation to make AI profitable, Bercovici said, since it's still an emerging field.

Related Topics: #AI #reinforcement learning #OpenAI #Anthropic #Gemini #AGI #latency #compute spend #coding agents

Can the AI industry absorb a surge‑pricing model? The discussion at VentureBeat’s AI Impact stop highlighted a growing capacity crunch, where latency spikes and cloud lock‑in threaten to curb rapid deployment. Roughly ninety percent of software now emerges from coding agents, according to Bercovici, suggesting agents have essentially come of age.

Yet reinforcement learning, the latest focus among labs such as OpenAI, Anthropic and Gemini, adds another layer of complexity to scaling efforts. Costs are climbing, and the analogy to Uber’s surge pricing underscores the immediacy of market pressures. Unclear whether pricing mechanisms will stabilize or exacerbate the strain on infrastructure.

Organizations may also face vendor lock‑in, complicating migration to alternative cloud providers. Meanwhile, developers confront higher latency, which could erode user experience if not managed. It's a shift toward agent‑generated code that is undeniable, but the long‑term impact on productivity and budget remains uncertain.

Stakeholders will need to balance innovation with practical limits as they navigate this evolving terrain.

Further Reading

Related Articles

Common Questions Answered

Why is the AI capacity crunch causing latency to increase for developers?

The capacity crunch stems from coding agents operating near their limits, which slows response times. As agents generate up to 90% of code, the heightened demand strains compute resources, leading to noticeable latency spikes.

How does the rise of coding agents affect cloud bills for companies?

With autonomous agents producing the majority of software, compute usage has surged, inflating cloud expenses. Companies that previously relied on human coders now see higher spend due to the intensive processing required by these agents.

What role does reinforcement learning play in the current AI scaling challenges?

Reinforcement learning is being explored by labs like OpenAI, Anthropic, and Gemini as a way to unify training and inference workflows. However, its integration adds complexity to scaling efforts, contributing to the overall capacity crunch.

Is the claim that 90% of software is now generated by coding agents accurate?

According to Bercovici, in certain workflows up to 90% of code is churned out by autonomous agents. This figure highlights the rapid adoption of agents, though it may vary across different development environments.