Urgent.News

What's breaking now, across thousands of outlets.

AI

I Love AI. I Still Don’t Want It in Everything.

I love AI. I use it constantly, I experiment with new models, I build things with it, and I genuinely think there are a lot of places where it can make software better. But I am also getting more and more tired of seeing AI being forced into literally everything. There’s a big difference between adding AI because it improves the product and adding AI because apparently every app now needs a…

Love for AI is evident, yet not all AI features are desirable. The author uses AI frequently for various purposes and believes it can enhance software. However, they are increasingly annoyed by AI being added to every app without a clear purpose. AI should only be included if it genuinely improves the product, rather than simply adding a "sparkle button" for the sake of it.

The author still incorporates AI in their personal projects as they appreciate its ability to streamline tedious tasks and improve efficiency. They started considering whether an app still functions perfectly without the AI feature. Offering users the ability to completely disable AI is preferred, eliminating any hints or prompts to use it.

The AI feature should seamlessly blend in without drawing attention. A clear question to ask when deciding on AI features is: What does AI actually improve in this context? If the answer isn't evident, it may not be necessary. Traditional software often performs well without AI. Not every problem requires probabilistic reasoning, natural language processing, or other AI capabilities.

Sometimes a simple filter is sufficient. Opting for traditional deterministic software can be beneficial as it's faster, predictable, and doesn't rely on external services or generate hallucinations. Optional AI can potentially enhance the AI features themselves. Users are forced to think more critically about the actual value AI brings.

The author is weary of projects hastily incorporating AI as an identity. Not every product needs an AI focus. It's essential to ask: Does AI truly add value to this product? Sometimes the answer might be no. The author remains excited about AI advancements but doesn't want AI to become a standard feature in every product, similar to adding Bluetooth to a toaster.

The ultimate goal should still be to create a high-quality tool. While AI can sometimes improve software, knowing when not to implement it is equally important.

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

Mistral Small 3.2 Lands With Sharper Function Calling and a 128K Context Window

Mistral just pushed Mistral Small 3.2 (24B) to Hugging Face, and the timing is interesting. While OpenAI keeps nudging GPT-4o mini's pricing and Anthropic gates Claude Haiku behind API quotas, Mistral…

  • Mistral Small 3.2 model introduced with 24 billion parameters and 128K context window.
  • Enhanced function calling with v3 tool schema format, 78% accuracy improvement.
  • Open-weight model allows fine-tuning and deployment without additional fees.

The Dependency the Agent Invented: A Myth-Busting FAQ

Consider a reconstructed incident that shows up when teams treat an agentic loop as an architecture review. A backend pair watched their coding agent close six steps with a confident summary and a…

  • A successful loop only shows code consistency, not external dependencies
  • Multi-step traces indicate scenery familiarity, not dissenting opinions
  • Zero exit code means process completion, not external object mutation

I let my own 31B model take over development of the thing running it

Two things I build: Flash , a local coding agent that runs in your terminal. Flash Onyx , the model it runs on. Gemma4 with a system prompt and sampling baked in.

  • Flash Onyx, a 31B model, controls its own development.
  • System prompt and sampling enable operation on Gemma4 architecture.
  • Self-policing mechanism ensures model accountability.

More from Saturday 5 September →