Urgent.News

What's breaking now, across thousands of outlets.

AI

Artificial intelligence programs create applications within hours

AI models can now generate functional software applications from scratch in hours, leading to projects like the open-source PhotoCraft. However, researchers point out that rapidly generated applications often lack proper security controls and contain significant code volume challenges. Brandon Thomas, head of the company ArtCraft, created the first version of PhotoCraft by giving simple…

AI models now possess the capability to generate fully functional software applications from scratch within mere hours. This breakthrough has led to initiatives such as the open-source project PhotoCraft. Nevertheless, experts have highlighted that hastily produced applications often fall short in terms of security measures and are plagued by substantial code volumes.

Brandon Thomas, the founder of ArtCraft, developed the initial iteration of PhotoCraft by providing straightforward instructions to an AI program over the course of approximately 24 hours. PhotoCraft, an open-source application, competes directly with Adobe Photoshop, boasting an appearance and operation virtually indistinguishable from its high-end counterpart.

PCMag's photography reviewer, Jim Fisher, conducted tests on both PhotoCraft and LightCraft, finding them to be impressively similar to Adobe's offerings, albeit lacking certain advanced functionalities. Concurrently, WordCraft, the equivalent application to Microsoft Word within ArtCraft's portfolio, is currently around 60 percent complete for practical use, according to its developers.

Ivan Kutskir, the creator of Photopea, emphasized that his program was not developed using AI, with its foundation dating back to 2012. In contrast, Anthropic has temporarily halted access to its Fable 5 model due to export control regulations pertaining to cybersecurity concerns.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

How to Secure an AI Agent That Can Access External Tools

Your agent can read email, query databases, and call APIs. Here is how to make sure it only does what it should. When an AI Agent Can Do More Than Answer Questions Picture an internal assistant at a…

  • Implement principle of least privilege for AI agents.
  • Separate capability from authorization in tool usage.
  • Mitigate risks like prompt injection and unauthorized tool execution.

More from Thursday 29 October →