The Rise of AI Software Developers: Devin and Beyond
A new startup, Devin, has made waves this week by introducing an AI program that can perform tasks typically handled by well-compensated software engineers. While chatbots such as ChatGPT and Gemini have already demonstrated the ability to generate code, Devin takes it a step further by planning problem-solving strategies, writing the code, and then testing and implementing the solutions.
Adept’s Action Transformer: Automating Useful Chores
When tasked with testing Meta’s Adept Action Transformer, an open-source program designed to automate useful chores by interacting with a user’s computer and the internet, Devin showcased its capabilities as an “AI software developer.” This development highlights the growing potential for AI to streamline and enhance various aspects of our digital lives.
SIMA: The Generalist AI Gamer
Google DeepMind’s SIMA, a groundbreaking AI agent, has recently demonstrated its prowess in playing video games, including the eccentric title Goat Simulator 3. By observing human players, SIMA has learned to execute over 600 complex tasks, such as felling a tree or destroying an asteroid. Remarkably, SIMA can successfully perform many of these actions even in unfamiliar game environments, earning it the title of a “generalist.”
“Making them more precise is something that we’re actively working on,” Tim Harley, a research scientist at Google DeepMind, told me. “We’ve got various ideas.”
The Future of AI Agents: From Video Games to Real-World Applications
While video games serve as an ideal testing ground for developing and refining AI agents, it is likely that Google harbors aspirations for these agents to eventually venture beyond the gaming realm. Potential applications could include assisting users with web-based tasks or operating software on their behalf. As these agents continue to evolve and improve, their impact on our daily lives is set to grow exponentially.
Google DeepMind’s Vision: Combining Large Language Models and AI Gaming Expertise
Demis Hassabis, the CEO of Google DeepMind, has recently shared his vision for the future of AI agents. By merging large language models with the company’s extensive experience in training AI programs to play video games, Hassabis aims to create more capable and reliable agents.
“This definitely is a huge area. We’re investing heavily in that direction, and I imagine others are as well.” Hassabis said. “It will be a step change in capabilities of these types of systems—when they start becoming more agent-like.”
As we look ahead, it is clear that the development of AI agents will continue to be a focal point for many companies in the coming months. With the rapid advancements in this field, we can anticipate a significant shift in the capabilities of these systems as they become increasingly agent-like, paving the way for a future where AI plays an even more integral role in our lives.
5 Comments
Guess we better start brushing up on our small talk, AI’s not waiting for us to catch up!
So, AI’s about to chat us up next level, huh? Interesting.
AI’s leveling up its game; we’re practically in the future now, talking to robots like it’s nothing.
Seems like we might need to up our conversation game, AI’s about to steal the spotlight!
AI just keeps getting chattier, doesn’t it? Watch out world!