GoKawiil
Tech News
A software developer recounts that their business partner used an AI coding assistant to build an entire payments backend module in a single day, then submitted a 5,236-line pull request without testing it. The developer found the endpoints didn't actually work and that documentation was riddled with obvious AI-generated filler, while missing critical details like where to obtain API keys.
Vibe coding, the practice of using AI language models to write software with minimal manual coding, is drawing scrutiny after Georgia Tech researchers linked dozens of security vulnerabilities directly to AI-generated code in just a three-month sample. Despite these risks, a survey of over 1,100 professional programmers found that 72 percent use AI coding tools daily, with AI-generated or -assisted code making up 42 percent of their codebases and expected to exceed half by next year.
A growing number of individual investors are using generative AI to write custom trading scripts and delegate portfolio decisions to automated agents, effectively mimicking strategies once reserved for professional quant funds. This trend, sometimes called 'vibe-coding,' lets non-programmers describe a trading idea in plain language and have AI translate it into executable code.
Google is reportedly preparing to launch Gemini 3.8 Flash imminently, a model DeepMind's internal testing found outperforms Anthropic's Opus on coding tasks. The release would follow Gemini 3.7 Flash, launched just weeks earlier, and comes as Google shelves plans for a separate Gemini 3.5 Pro model, pushing the next Pro-tier release to Gemini 4.0.
An Entrepreneur contributor describes using vibe coding, building software through natural-language prompts to AI, to run their solo business more efficiently. They report automating repetitive workflows, launching products faster and creating interactive marketing tools without hiring developers, turning tasks that once took weeks into hours-long projects.
Meta has launched Pocket in the US, a mobile app built by the former Gizmo team that lets users create interactive game prototypes—called 'gizmos'—simply by typing text prompts, with no coding or code visibility involved. Finished creations can be shared in a TikTok-style feed with likes, comments and reposts, but users have no way to export or use them outside the app.
Paul Sutter, an astrophysicist and science communicator, wrote in Nautilus that he used AI 'vibe coding' to build a faster, more sophisticated update to his void-detection algorithm. Ten minutes into presenting it to collaborators, a colleague flagged that its handling of survey edges was fundamentally wrong, meaning every result built on it was also incorrect.
Jotform's CEO, a 20-year software veteran, describes how AI tools that generate functional code from plain-language prompts are letting non-programmers build working prototypes quickly. He argues this approach is especially powerful for creating minimum viable products and automating repetitive coding tasks, though it still requires oversight from someone who understands the underlying logic.
Meta has rolled out its Pocket app to all users, letting anyone create simple games, interactive gizmos, and other small digital experiences by typing a description instead of writing code. A ZDNET writer tested it by requesting a cookie-baking game for increasingly impatient customers, and the app generated a working, if quite difficult, version in under a minute.