🍭 More Meta Ads + "Vibe Working" 🤔

Microsoft + Meta both drop news! More ways into our data? Oh great......

Good morning. My laptop just asked for PTO because Copilot’s doing its job now.

Denied.
We have vibes to ship.

Let’s dig in.

🍭 What’s Cookin’:

  • Meta found a new way in to your data

  • Microsoft is shilling “vibe-working”

  • And Everything Else

ToolBox™
🙈 5 New drops in AI You Need to Know About

→ Text-to-video leveled up: longer clips, stronger physics, and prompt control that finally feels film-maker friendly.

→ An AI code editor that orchestrates multiple agents in parallel so refactors, tests, and docs ship together.

⌨️ Cursor 1.7

→ Agent autocomplete with project-aware hooks and team rules; your repo’s norms, enforced by the IDE.

🦁 Ask Brave

→ Private, source-backed answers from the browser company that won’t sell your soul (or your search).

🗓️ Everyday

→ Tell it what you need in plain English and it executes across apps—tasks, calendar, and workflows included.

Meta/Facebook
🗞️ Meta Will Soon Use Your AI Chats To Tune Ads.

The Bite: Starting on December 16th, anything you tell Meta AI, text, voice, even Ray-Ban glasses snaps, will be used to shape what shows up in your feed.

Ads, posts, the works.
No opt-out button (other than ghosting the AI).

Notices start rolling Oct 7, with UK/EU/S.
Korea exempt thanks to regulators.

Sensitive stuff like health or politics is off-limits… allegedly.

Snacks:

  • Marketer jackpot: Real-time intent beats Likes and follows.

  • No clean escape: Only way out is not using Meta AI.

  • Cross-app creep: Link accounts and signals flow across FB/IG/WhatsApp.

  • Hardware hook: Ray-Bans now double as ad-fuel cameras.

  • Geo wall: GDPR shielded Europe; Asia’s got PIPC.

Why it bites: For Meta, this is the fattest ad-signal upgrade since the Like button. Explicit intent at scale = billions in new ad juice, right when Apple throttled old targeting.

For users, it’s confession booth → ad auction, with no “pay-to-escape” tier in sight (though YouTube/Apple already showed it could exist).

We hope Meta gets there, but until regulators twist arms, you’re just raw signal.

And shoutout to GDPR, the rare time EU red tape feels like a blessing.

UK aside: can I get arrested for being rude to my AI therapist now? 🙈

Which image is real?

One's real. One's AI

Login or Subscribe to participate in polls.

Voice Agents
📞 Build Voice AI Agents with Telnyx

Most AI agents fail under real-world pressure with lag, unclear audio and scaling issues.

Telnyx built the infrastructure for real-time conversations.

• Sub 200ms latency for natural interactions

• HD voice with global telephony in 60+ countries

• Transparent $0.05 per minute pricing

Steal This Prompt
👻 Liminal Spaces, Infinite Vibes

Summon eerie “in-between” scenes—after-hours malls, never-ending hallways, buzzing fluorescents, that feel like déjà vu you can’t place, all tuned by your own location, lighting, and time-of-day inputs.

  1. Click this link for the prompt

  2. Copy the prompt

  3. Paste into GPT

  4. Watch your brain say “I’ve been here before,” but it hasn’t.


    🍭 Try the prompt on Snack Prompt 🍭

Everything Else
🧠 You Need to Know

🧩 OpenAI taps Samsung and SK Hynix for Stargate memory

🏠 Google unveils AI-powered Nest cams and a new doorbell

🧑‍💻 Salesforce launches Agentforce Vibes for “vibe coding”

🎮 Epic: Apple’s revised iOS install flow cuts drop-offs 60%

📚 Wikimedia project makes Wikipedia data easier for AI

Microsoft Copilot
🤖 Microsoft’s “Vibe Working” Pitches Agentic Office.

The Bite: Microsoft just put a bow on agentic workflows in Office: “vibe working.”

Behind the cringe name is a legit shift, Copilot can now run Agent Mode inside Word and Excel, plus a new Office Agent in Copilot chat.

Instead of one-and-done prompts, you can actually coach the AI through drafts, revisions, and follow-ups.

Sounds small, but because Microsoft owns the Windows plumbing, these agents can plug into the real guts of files, data, and permissions in a way Google can only dream of.

The big question:
will schools, colleges, and government, where Windows is basically oxygen, actually use it?

Snacks:

  • 1The Fine: CA lawyer nailed for 21 fake quotes → $10k.

  • The Rule: New mandate: every court must ban AI or set policy by Dec 15.

  • Not Isolated: NY’s Avianca case also saw sanctions for bogus ChatGPT citings.

  • Pattern Building: Puerto Rico lawyer hit with $24.4k over 55 bad citations.

Why it Bites: Courts are stepping up where politicians won’t.

By slapping fines and mandating rules, they’re sending a clear message:
use AI, but own the output.

That means labeling when something’s fully AI-written, keeping sensitive info out of public tools, and, most importantly, checking every quote like your license depends on it.

Because now? It does.

The trust math is simple: clear rules + real penalties = less AI slop, more accountability.

And if it takes a few $10k bonks to get everyone verifying their receipts, so be it.

can i just trade monkey nfts in the metaverse mom?