Veo 3.1 Just Launched. Google Pushes AI Video Closer to Real Filmmaking.

Plus: OpenAI’s adult mode, Anthropic’s Haiku 4.5, and AI now outwrites humans

Google has released Veo 3.1, an update to its AI video model that adds sound, better editing tools, and more control for creators. The model powers Flow, Google’s AI video platform, which has already seen more than 275 million videos made since launch.

What’s new in Veo 3.1:

• Audio in motion: Veo now adds natural sound to its videos, helping scenes feel more alive and complete.
• Creative control: Users can guide videos with reference images, connect scenes frame by frame, and extend clips up to a minute-long.
• Editing tools: Flow now lets users insert or remove objects in a scene, adjusting lighting and texture to keep the shot consistent.
• Access: Veo 3.1 is available in Flow, Gemini, and Vertex AI, so both developers and creators can use it.

Veo 3.1 arrives at a time when AI video models are competing to move beyond visuals. OpenAI’s Sora 2 has been in the spotlight, but Google is taking a quieter route by refining what already works. Instead of trying to shock people with spectacle, it’s building tools that feel like a full studio, only more accessible.

OpenAI CEO Sam Altman has confirmed that ChatGPT will soon support erotic conversations for verified adults. The change is part of a larger effort to make the chatbot sound more natural and human in its tone. The update is expected to roll out in December, alongside new personality features that give users more control over how ChatGPT behaves.

What’s changing:

• Adult mode: Verified users will be able to access an erotic chat setting under OpenAI’s new “treat adult users like adults” policy.
• Safety checks: The feature will rely on OpenAI’s new age-prediction system, with ID verification for any user misclassified as a minor.
• Wider updates: A new version of ChatGPT will make the assistant friendlier, more conversational, and expressive.
• Context: OpenAI is still under scrutiny for earlier mental health concerns linked to users forming emotional bonds with its AI models.

Altman says the company has “mitigated serious mental health issues” through improved safeguards and expert oversight. Researchers remain uneasy about how the feature could affect vulnerable users. Erotic AI chats have already driven massive engagement on apps like Character.AI, where users spend up to two hours a day talking to bots. With 800 million weekly users, OpenAI’s entry could redefine that market overnight.

Anthropic has released Claude Haiku 4.5, the newest and smallest member of its model family. The update brings performance that rivals much larger systems like Claude Sonnet 4 and even OpenAI’s GPT-5, while running at one-third the cost and more than twice the speed.

What to know:

• Compact and capable: Haiku 4.5 matches Sonnet 4 in coding accuracy, according to the SWE-bench Verified benchmark.
• Faster answers: The model delivers quick, concise responses ideal for mobile use and high-volume tasks.
• Cost advantage: Free users get it by default, while paid users can access it at a fraction of Sonnet’s price.
• Teamwork: Haiku 4.5 can handle subtasks while Sonnet 4.5 manages planning, allowing both models to work together on complex workflows.

Anthropic says the model “punches above its weight,” offering smaller businesses and developers a low-cost way to build AI systems that can think and code efficiently. For companies tracking the SWE-bench numbers, Haiku’s results suggest a new era of lightweight models that can stand toe-to-toe with giants and fast enough for real-time coding, cheap enough to scale across every desk.

A new study from Graphite found that AI-written articles briefly passed human-written ones on the web in late 2024. Since then, the growth has slowed, and the internet is now split almost evenly between the two.

What the study found:

• Large sample: Researchers analyzed 65,000 English-language articles published between 2020 and 2025.
• Rapid growth: AI-written content surged after ChatGPT’s launch and overtook human output by November 2024.
• Leveling off: The share of AI articles has stayed mostly the same since then.
• Likely reason: Many publishers found that AI-written pieces do not perform well in search results.

The rise of AI content has changed how information spreads online. Tools can produce millions of articles quickly, but much of that writing still fails to reach readers. Human writing continues to hold value, not because it’s slower or harder, but because people still trust the voice behind it.

Invest where the future of AI begins. AI agents will drive tomorrow’s economy — the foundation is being built today.

⚠️ Final days. The bonus-share offer ends October 17, 2025. Act now to get up to 25% extra shares before the window closes.

Why GenAI.works:

• Infrastructure for agentic AI systems

• Trusted by 200+ companies / 13M+ global community

• Bootstrapped to $3M ARR

👉 Put your capital where tomorrow’s economy starts.

Secure My Bonus Now

Reg CF offering via DealMaker Securities. Investing is risky.