For more than a decade, tech companies were in desperate need for software engineers. They needed more coders. But today’s college graduates with computer science and computer engineering degrees are facing a tough job market.1 Many factors are at play, but AI’s ability and potential to write code may be a huge reason for the drop in demand for coding jobs—for humans at least.
According to new research by MIT, 82% of software engineers (coders) said they regularly use AI tools to help them code.2 Many of these engineers are being encouraged to utilize AI by their CEOs who tout the ability of LLMs (large language models) to generate code. Tech’s top CEOs anticipate AI will be writing the majority of code within mere months.3
While ambitious, those expectations have not yet come to fruition. While AI coding tools have been shown to create good short code snippets, they lack the critical thinking needed to generate complex code that solves real-world problems.3 AI code is prone to producing errors and when asked to test code can sometimes get stuck in a loop.3
AI coding tools, while helpful in many scenarios, still need human oversight in most cases. “Long-horizon code planning” involves a critical understanding of the ways in which code fits into complex systems, how end users will use it, and what the consequences might be beyond the code itself—all skills still only held by humans to the extent they’re needed in coding.2 AI coding tools also haven’t totally won the trust of engineers just yet.3
Because of AI’s current limitations with coding, humans have been utilizing it to assist them with menial, repetitive, simple tasks that many say saves them precious time throughout their workweek.3 As AI continues to be used to code, AI developers will continue to improve these tools to make them more powerful. As AI is in its infancy, its abilities and efficiencies will undoubtedly improve over time. Developers will train AI on a wider variety and complexity of coding tasks, like testing code, maintaining code, and interacting with human collaborators.
AI is only just now at the very early stages of what it may be capable of achieving within coding. Since the ChatGPT breakthrough in 2022 rocked the tech world, the AI race has accelerated at breakneck pace. Big tech has buoyed the markets and driven most of its growth since then, all in the name of making AI bigger, better, and more pervasive than ever.
The TrueShares Technology, AI, and Deep Learning ETF (LRNZ) is designed to navigate the early, hyper-growth stages in this unique technology space. LRNZ generally holds 20-30 positions representing companies we believe possess innovative AI and deep learning solutions that represent a distinct competitive advantage in a particular industry. AI, as evidenced by its ability and potential to code, is only just getting started. LRNZ can act as an accessible entrypoint for avoiding market FOMO.