Amazon founder Jeff Bezos recently argued that software engineers should not fear artificial intelligence at all. Comparing AI tools to “a bulldozer replacing a shovel”, Bezos suggested that developers
who learn to work with AI could become dramatically more productive and efficient.
But at almost the same time, executives at Anthropic — one of the world’s leading AI companies — warned that AI could fundamentally alter white-collar work itself, especially in industries heavily dependent on repetitive execution tasks like coding, testing and documentation.
For India, which has 4.3 to 5 million software engineers and IT professionals, much of the ecosystem depends on large workforces handling repetitive programming tasks for global companies. Now, generative AI is beginning to automate many of those very functions.
Why Jeff Bezos Thinks Engineers Should Embrace AI
Jeff Bezos belongs to a growing camp of tech leaders who believe AI will ultimately make engineers more powerful rather than irrelevant.
Speaking with CNBC from Blue Origin’s rocket facility in Florida on Wednesday, the Amazon founder compared AI to having a “bulldozer rather than a shovel to dig a big hole”.
“If you’ve been digging out a basement for your house with a shovel and somebody’s about to hand you a bulldozer, you should be so happy,” he said about software engineers. “It’s going to elevate all these people, he added. “We are going to have so much productivity in our economy”.
The argument is rooted in productivity. AI-powered coding assistants can already write code, fix bugs, generate documentation, summarise software systems and automate repetitive programming work within seconds. Tasks that once took hours can increasingly be completed in minutes.
From Bezos’ perspective, this does not necessarily destroy engineering jobs. Instead, it changes the nature of the work itself.
Rather than spending time on repetitive coding or debugging, engineers may increasingly focus on high-level thinking, system architecture, problem-solving and product design. AI, in this vision, acts less like a replacement and more like an intelligent collaborator.
The comparison to a bulldozer replacing a shovel reflects the belief that technological revolutions historically improve productivity even if they disrupt older workflows. But not everyone in the AI industry sounds equally optimistic.
Why Anthropic’s Warning Matters
Anthropic’s leadership has increasingly suggested that AI may significantly reduce the amount of “execution work” humans perform across white-collar professions.
CFO Krishna Rao said AI now writes 90% of Anthropic’s code on a podcast on Wednesday. He described the workplace inside Anthropic where AI systems increasingly handle the execution work from software engineering to financial reporting, while human work involves oversight, judgment, and strategy.
“We’ve hired a lot more people because of that,” Rao said, adding that he sees Claude as a productivity ‘accelerant’.”
This could mean developers spending less time writing code line by line and more time checking AI-generated outputs for errors, reliability and security risks. That shift may sound subtle, but it could transform how the tech workforce operates.
Historically, entry-level engineering jobs have depended heavily on repetitive hands-on tasks. Junior developers often begin with software testing, debugging, documentation, backend maintenance and routine coding support.
These are exactly the kinds of functions AI is becoming increasingly capable of automating.
Anthropic’s own research has also raised concerns around how AI coding assistance may affect long-term skill development. It said developers using AI scored 17% lower on post-task quizzes covering debugging and conceptual reasoning compared to those who coded by hand.
While developers using AI tools may complete tasks faster, researchers are still studying whether heavy dependence on AI could weaken deeper learning and independent problem-solving abilities over time.
In a study published in April, researchers in the US and UK found that people who spent just 10 minutes using AI to help solve math or reading-comprehension problems, their own “unaided performance” on some problems diminished.
The people who took help from AI not only fared worse than a second group who had worked without AI assistance, but they also gave up on challenging problems more quickly, the study highlighted.
How AI Is Reshaping Software Engineering
AI coding assistants such as GitHub Copilot, Claude and other generative AI developer tools are already being integrated into software workflows across technology companies worldwide.
These systems can generate code, identify vulnerabilities, automate testing, explain unfamiliar programming languages and even suggest improvements in software architecture. Increasingly, software engineering is evolving from manually writing every line of code towards managing intelligent systems that assist with development.
This is creating demand for what many companies now call “AI-native engineers”; professionals who know how to work alongside AI systems effectively rather than compete against them.
“AI-native engineers are developers who work alongside AI from the very beginning instead of doing everything manually. Rather than writing every single line of code themselves, they guide AI tools, review outputs, fix errors and focus more on building stable and efficient systems,” said Aayushi Sharma, senior business analyst at an IT company in Noida.
As AI tools become more sophisticated, engineering value may depend less on memorising coding syntax and more on critical thinking, systems design, cybersecurity awareness and the ability to evaluate whether AI-generated outputs are reliable.
In practical terms, developers may spend less time coding and more time supervising AI-driven workflows.
How This Will Impact Jobs In India?
For India, the AI transition could prove far more disruptive than for Silicon Valley itself. India’s IT services economy was built on scale. The country became a global technology outsourcing powerhouse by supplying large engineering workforces capable of handling software maintenance, migration, support and testing operations for international companies.
The Indian IT-BPM industry contributes over $250 billion annually to the economy and employs millions directly. Companies such as Infosys, TCS and Wipro expanded rapidly using labour-intensive service models.
But generative AI directly threatens many of the repetitive workflows that powered this growth.
If AI-assisted engineers can complete tasks once handled by multiple junior employees, companies may eventually reduce entry-level hiring needs. Industry estimates show that junior IT roles have declined by 20-25% due to the adoption of AI.
Many technology companies are already prioritising candidates with AI-related skills over traditional programming capabilities alone. For decades, IT majors like TCS and Wipro and large outsourcing firms hired tens of thousands of freshers from engineering colleges to perform basic manual coding. Today, hiring has become highly targeted and cautious, with firms favouring experienced professionals with at least 4+ years of experience who require less initial training.
The fear is not necessarily that software engineering will disappear entirely. Instead, the bigger concern is that the bottom rung of the career ladder could shrink sharply.
Why The Engineering Education System Faces A Major Test
The rise of AI is also exposing weaknesses within India’s engineering education ecosystem. Every year, India produces roughly 1.5 million engineering graduates. But only 45% of them meet specific skill requirements and standards demanded by modern industries. Yet many colleges still focus heavily on theoretical coding, rote learning and outdated curriculum structures.
With AI automating routine programming tasks, educational priorities may need to shift dramatically towards problem-solving, systems thinking, AI management, cybersecurity and interdisciplinary learning.
Highly skilled engineers capable of building advanced AI systems could become significantly more valuable and command higher salaries. Meanwhile, workers performing repetitive coding and support functions may face falling demand.
This could create a divided technology economy where AI-native engineers thrive while traditional coders struggle to adapt.
Can India Move To AI Product Creation?
Despite the disruption risks, many experts believe India still has a major opportunity to benefit from the AI transition.
The country already possesses strong digital infrastructure, a massive developer ecosystem and a rapidly expanding AI start-up landscape. Indian firms are increasingly integrating AI into fintech, healthcare, logistics, enterprise software and customer service operations.
If India successfully moves beyond low-cost outsourcing and begins building globally competitive AI products and platforms, the country could remain a major technology power in the AI era.
But the transition may require speed, reskilling and large-scale adaptation.
Countries and companies that learn to integrate AI effectively into engineering workflows are likely to dominate the next phase of global technology growth.
