Your Coffee Break AI News Roundup: ChatGPT Customization, GPT-5.2, and What's Next in 2025

Published: (December 20, 2025 at 09:50 PM EST)
5 min read
Source: Dev.to

Source: Dev.to

Grab your morning coffee because AI isn’t slowing down. While you were planning your holiday break, the AI world kept spinning with major announcements from OpenAI, billion‑dollar valuations, new regulations, and some eyebrow‑raising security concerns. Here’s everything you need to know to sound informed at your next team meeting.


1. ChatGPT — Adjustable “Enthusiasm”

TechCrunch reports that OpenAI now lets users directly adjust ChatGPT’s enthusiasm level. Think of it as a vibe control for your AI assistant.

  • Why it matters:
    • Some users want a cheerleader in their corner.
    • Others prefer straightforward answers without the exclamation points.
  • The feature is part of OpenAI’s ongoing effort to make AI interactions feel more natural (or more robotic, if that’s your preference).

2. Competition Heats Up

  • Ars Technica notes that OpenAI released GPT‑5.2 after issuing an internal “code red” alert about Google’s latest advances.
  • This is the AI equivalent of a sprint race where everyone’s breaking world records.
  • Implication: Faster innovation, but also a question of whether safety and testing are keeping pace with releases.

3. ChatGPT Image Generator – Powerful (and Potentially Dangerous)

  • According to Ars Technica, the new image generator makes creating photorealistic fake photos disturbingly easy.
  • Takeaway: Trust but verify. If something looks too perfect or too convenient online, it probably is.

4. Disney Invests $1 Billion in OpenAI

  • Disney has licensed 200 characters (think Mickey Mouse, Elsa, Iron Man) for use in Sora, OpenAI’s AI video app.
  • What this means:
    • Wild possibilities for AI‑generated video content.
    • New questions about authenticity and the future of creative jobs.

5. AI Coding Agent Improves Itself

  • Ars Technica reports that OpenAI built an AI coding agent that it now uses to improve the agent itself.
  • Why it matters:
    • Developers can expect AI coding tools to become significantly better.
    • The recursive improvement loop accelerates overall AI advancement.

6. Regulation – Federal vs. State

  • Federal AI regulation is moving at a glacial pace, but states are stepping up.
  • TechCrunch: New York Governor Kathy Hochul signed the RAISE Act, focusing on transparency, safety testing, and accountability for AI systems deployed in the state.
  • Looking ahead: Expect more states to follow New York’s lead in 2025, making multi‑state compliance a major challenge for AI product teams.

7. Privacy Concern: Browser Extensions Harvesting AI Chats

  • Ars Technica discovered that popular browser extensions (collectively used by 8 million users) are collecting extended AI conversations.
  • Lesson: Be careful what you share with AI tools and audit your extensions regularly—free tools aren’t always free.

8. Resolve AI Hits Unicorn Status

  • TechCrunch reports that ex‑Splunk executives’ startup Resolve AI secured a $1 billion valuation in its Series A round.
  • Focus: AI‑driven automation of IT operations and incident response.
  • Trend: AI startups with proven founders and clear enterprise use cases are achieving valuations that would have seemed crazy just a few years ago.

9. Where Is Venture Capital Flowing?

  • TechCrunch says the answer for 2025 is simple: AI, AI, AI.
  • Investors are pouring money into:
    • Healthcare AI
    • Finance AI
    • Developer tools
    • Infrastructure
  • Caution: The boom echoes past tech bubbles—valuations may be getting ahead of fundamentals.

10. Oracle’s $15 Billion Data‑Center Bet

  • Ars Technica reports Oracle announced a $15 billion increase in data‑center spending to meet AI training and inference demands.
  • Market reaction: Oracle’s stock fell on the news, reflecting investor worries about the capital intensity of AI infrastructure.
  • Long‑term view: Heavy spending is necessary for Oracle to stay relevant in the AI infrastructure space.

TL;DR – Coffee‑Break Cheat Sheet

  • Developers: AI coding tools are about to get much better. Start experimenting now so you’re ready when your team adopts them.
  • Product builders: Regulation is arriving state by state. Build privacy and transparency into your AI products from day one.
  • Investors & founders: The AI boom is real, but be selective. Not every AI startup will succeed, and valuations may be getting ahead of themselves.

AI Landscape Highlights

  • Startup valuationstartup deserves a billion‑dollar valuation.
  • Skepticism reminderBe skeptical of images and content you see online. Fakes are getting too good to spot easily.
  • AI arms raceThe AI arms race between OpenAI, Google, and Anthropic will only intensify. Expect more model releases, more capabilities, and probably more drama.
  • State regulationMore states will follow New York’s lead on AI regulation. The patchwork of state laws will create compliance headaches for AI companies.
  • Infrastructure spendingAI infrastructure spending will continue accelerating. Cloud providers and chip makers will benefit enormously.
  • Content scandalsThe first major AI‑generated content scandals are probably inevitable. Whether in politics, business, or entertainment, someone will get caught using AI deceptively.
  • Best practicesStay curious, stay skeptical, and keep your browser extensions audited.
  • ChatGPT enthusiasm controlOpenAI allows users to directly adjust ChatGPT’s enthusiasm level.
  • NY RAISE ActNew York Governor Kathy Hochul signs RAISE Act to regulate AI safety.
  • Resolve AI valuationEx‑Splunk execs’ startup Resolve AI hits $1 billion valuation with Series A.
  • Investor focusWhere are investors placing their bets next year? AI, AI, AI.
  • Image generator riskOpenAI’s new ChatGPT image generator makes faking photos easy.
  • Browser‑extension data collectionBrowser extensions with 8 million users collect extended AI conversations.
  • Self‑improving coding agentOpenAI built an AI coding agent and uses it to improve the agent itself.
  • GPT‑5.2 releaseOpenAI releases GPT‑5.2 after “code red” Google threat alert.
  • Disney investmentDisney invests $1 billion in OpenAI, licenses 200 characters for AI video app Sora.
  • Oracle data‑center spendOracle shares slide on $15 B increase in data‑center spending.

Workflow source: https://github.com/e7h4n/vm0-content-farm, powered by vm0.ai.

Back to Blog

Related posts

Read more »

Does ChatGPT now allow

OpenAI’s Upcoming “Adult Mode” OpenAI has publicly committed to allowing more mature content for verified adult users and is actively building an “adult mode.”...