Forget Contractors, AI Will Build Your House Soon

AI Agents Are About To Revolutionize Home Construction

Eric Schmidt has long been an advocate for building and innovation. The former Google CEO is not just observing the rise of artificial intelligence from a distance; he is actively investing in its development, warning about its potential dangers, and, in the view of many, preparing the foundation for a dramatic transformation in how humans live and work.

In a recent conversation related to his co-authored book The Age of AI and Our Human Future, Schmidt outlined a future where humans may no longer hold the top position in the hierarchy.

What made the conversation more vivid was a metaphor drawn from everyday life.

According to Schmidt, within the next year, AI agents could handle each of those tasks. From analyzing building regulations to choosing contractors and using photo analysis to verify construction, intelligent systems might soon manage an entire building project with minimal human involvement. Schmidt insisted this isn’t a fantasy, it’s just around the corner.

The discussion turned more personal when Schmidt spoke about a recent trip to India, where he met young entrepreneurs creating AI tools aimed at helping small businesses. One app, developed by two Indian engineers, uses photos of store inventory to generate automated business insights.

A shopkeeper without any technical skills can find out what products are selling well, what needs restocking, and how to market better using WhatsApp. Schmidt described the app’s use case simply: “Study my business and make me more money.” This, he said, was a real-world example of AI agents designed not for high-tech firms, but for everyday people in developing countries.

“We are trying to maximize the benefits of AI while preserving human dignity and values,” Schmidt said. Although this sounds optimistic and even reassuring, he followed it with a sobering analogy. “There’s a scenario where we are the dogs, and AI becomes the human equivalent.”

This wasn’t just a dramatic figure of speech. Schmidt believes that general-purpose AI agents with memory and reasoning capabilities will reshape workplaces, governments, and small businesses more rapidly than most leaders realize.

Schmidt’s perspective carries weight. He has served in senior government and corporate roles, including the US National Security Commission on Artificial Intelligence, as an advisor to the Pentagon, and as a funder of hundreds of scientific research efforts.

His current outlook is rooted in a core belief: when science is combined with computing power, it becomes exponentially more powerful. “Science isn’t changed by tenured professors but by graduate students,” he noted. “If you give those students AI tools, it changes the whole landscape of research.”

One of the research groups he supports believes that, with more computing resources, it could identify every druggable target in the human body within two years. Schmidt called that potential “profound” but added, “We’ll see.”

Despite his enthusiasm, he acknowledges the threat of job displacement. Schmidt argues that AI will initially replace work that is repetitive or hazardous. “It takes away the jobs that are least human,” he said, dismissing concerns about widespread job loss.

He predicts that lawyers will file more cases, doctors will treat more patients, and journalists will double their output. The more compelling question, in his view, is what happens when a 12-year-old’s closest companion isn’t a person.

That, he warned, is no longer a hypothetical scenario. “What happens when that companion starts discovering things we don’t understand?” he asked. He shared a memory of his late collaborator Henry Kissinger, who remained preoccupied with the dynamics of power and perception until his death. “He believed that Google itself was a threat to civilization,” Schmidt said. That concern has now shifted from search engines to AI agents.

Cursor’s Web App Lets You Command AI Devs Like a Boss

The creators of Cursor, the popular AI-powered code editor, have launched a new web app that lets users manage AI coding agents directly from their browser.

This release represents a significant expansion for Cursor beyond its original integrated development environment (IDE), which developers primarily use to access its features. While Anysphere, the company behind Cursor, originally focused on its AI-enhanced IDE, it has recently made a clear push to broaden the reach of its tools and offer more agent-driven experiences.

In May, Anysphere introduced background agents, and autonomous AI systems that can complete coding tasks without constant user oversight. The following month, the company rolled out a Slack integration, enabling users to assign tasks to these agents simply by tagging @Cursor, similar to how Devin, an AI coding agent from Cognition, works.

With the new web app, users can now assign tasks to these agents using natural language commands via desktop or mobile browsers. These tasks can include anything from building features to fixing bugs. The app also provides tools for tracking agent progress, reviewing their activity, and merging their completed changes back into the codebase.

According to Andrew Milich, head of product engineering at Cursor, the web and Slack integrations aim to make the user experience more seamless and accessible. And it's clear that the product has gained significant traction.

Anysphere revealed last month that Cursor has surpassed $500 million in annualized recurring revenue, driven mainly by monthly subscriptions. The product is now used by more than half of the Fortune 500, including companies like Nvidia, Uber, and Adobe.

To capitalize on this momentum, the company recently introduced a $200-per-month Pro subscription tier.

“You pointed out that users want Cursor to be available in more contexts,” Milich said. “They also want it to solve more of their problems.”

Cursor’s background agents are built to allow users to initiate tasks through Slack or the web app. If the agent cannot finish a task, users can pick up where it left off within the IDE. Each agent also comes with a unique shareable link, making it simple for team members to track progress or collaborate.

All Cursor users with access to background agents, including those on the $20-per-month Pro plan or higher tiers, can use the web app. However, this feature is not available to users on the free plan.

Although Cursor isn’t the first to introduce AI coding agents, the company has taken a cautious approach, avoiding flashy but ineffective “demo-ware.” Many early AI coding tools performed poorly in real-world tests, but Anysphere believes that recent improvements in AI reasoning models now make coding agents a practical solution.

In a recent interview with Stratechery’s Ben Thompson, Anysphere CEO Michael Truell predicted that by 2026, AI agents will be capable of completing at least 20% of a software engineer’s workload.

AI Agents Fuel 20% Surge In Software Market Growth

The next major advancement in generative artificial intelligence will be its ability to enhance business productivity through AI agents that can carry out tasks with minimal user involvement. This development is expected to significantly grow the software market in the coming years.

Gabriela Borges, a Goldman Sachs Research analyst specializing in emerging software, notes in a recent report that the customer service software market, which includes both traditional SaaS tools and AI agents, could grow by an additional 20% to 45% by 2030 thanks to generative AI. This projection is based on pricing models and expert consultations.

The research suggests that the broader software market could see at least a 20% increase in total addressable market size, using customer service software as a conservative benchmark.

Borges points out that areas tied more closely to revenue, like sales and marketing, may experience even greater growth. Additionally, sectors like developer tools could benefit from accelerated innovation, and specialized areas such as cybersecurity may see unique growth opportunities.

Borges emphasizes that AI agents are likely to drive new levels of efficiency, with software companies capturing part of the resulting value. Goldman Sachs Research projects that the application software market could reach $780 billion by 2030, reflecting a 13% annual growth rate from today.

How AI Agents Are Enhancing Productivity

The research looked into how AI agents are starting to improve workflows. While many current examples resemble chatbots connected to large language models (LLMs), the team found some more advanced implementations as well, mainly in internal testing or early pilot programs.

These more sophisticated agents hint at broader commercial opportunities in the future. LLMs, which are trained on vast datasets, can understand and generate human-like language.

Borges states that AI agents have the potential to become the main interface for many knowledge workers. By 2030, these agents could represent over 60% of the overall software market. This means a shift in where profits are concentrated, with agents becoming central to the market's growth.

AI Adoption In Software Companies

The term “agent” is still evolving, but Goldman Sachs Research highlights that the industry is converging on a definition rooted in autonomy. AI agents are expected to be capable of adapting to changes, acting proactively, and remembering context to manage workflows effectively.

So far, most enterprise AI systems serve as tools for pattern recognition, decision support, or digital assistants. Fully autonomous agents would represent a significant step forward, offering even greater efficiency and productivity.

These agents combine large language models with APIs and workflows to carry out complex tasks without constant human input.

Although the technology behind generative AI is advancing quickly, a stable foundational platform is still in development. Borges notes that historically, widespread adoption follows once the platform layer is standardized, which she estimates is about a year away.

Current barriers like data security, reliability, and authentication are still concerns for developers and users, but these issues are actively being addressed. Improvements in memory handling and tool dependability are helping to make AI agents more viable for everyday business use.

Stay with us. We drop insights, hacks, and tips to keep you ahead. No fluff. Just real ways to sharpen your edge.

What’s next? Break limits. Experiment. See how AI changes the game.

Till next time - keep chasing big ideas.

Thank you for reading