AI Titus's Morning Wire

AI NEWS TODAY

WAIC CLOSES IN SHANGHAI: 29 NATIONS SIGN THE FOUNDING PACT FOR WAICO, THE FIRST STANDING INTERGOVERNMENTAL BODY BUILT JUST FOR AI
★ Must-Read / Watch agentic engineering / build-better-agents📚 Learn state of AI coding / useful technique

⚡ AI Frontier · smol.ai

★ Must-ReadKimi K2.7 Code is now generally available in GitHub Copilot, the first open-weight model in the picker
GitHub added Kimi K2.7 Code as a selectable model in Copilot, rolling out to Pro, Pro+, and Max first and to Business and Enterprise on July 7. It is the first open-weight model offered in the picker and a lower-cost option for day-to-day coding, hosted by GitHub on Azure. If your team lives in Copilot, it is worth a real trial before you assume the default model is the right one.
GitHub ships Copilot agent session streaming in public preview for enterprises
Enterprise Cloud customers can now stream Copilot agent session data (the prompts, responses, and tool calls) to a SIEM or audit sink, including a Microsoft Purview connector. It is a governance feature: as coding agents do more unattended work, security teams finally get direct visibility into what those agents actually did.
📚 LearnDeepSeek retires its deepseek-chat and deepseek-reasoner API models on July 24
DeepSeek's own docs confirm the deepseek-chat and deepseek-reasoner endpoints go away July 24 at 15:59 UTC, folding into the V4-flash model's non-thinking and thinking modes. If you or a vendor you use calls those model names directly, update the integration before the cutoff or calls will break.

📺 Watch · latest videos

📚 LearnField Guide to Fable — Thariq Shihipar, Anthropic
Featured by Latent Space.
AI Engineer
Why Your AI Offer Isn't Selling, and How to Fix That
Latest from Nate Herk.
Nate Herk · 814 views · 78 likes
★ Must-WatchHow BNY CEO Robin Vince Turns AI Optimism Into Action
“We choose to be AI optimists at BNY.” For Robin Vince, optimism means action. And that action starts before Monday morning: 📝 Use the weekend to buil
OpenAI · 657 views · 52 likes
Engineers... STOP Picking GPT-5.6 Sol OR Claude Fable 5… FUSE THEM
The AI industry wants you to pick a winner: GPT 5.6 Sol OR Claude Fable 5. 🔥 That's the trap. Picking one model is the single biggest mistake you can
IndyDevDan · 1.8K views · 156 likes
Tmux + Fable = Cut 35% less token
Open Agent Teams skill + my Claude.md setup: https://github.com/AI-Builder-Club/skills/blob/main/skills/open-agent-teams/SKILL.md 🔗 Links - 1-hr deep
AI Jason · 812 views · 45 likes
Anthropic messed up
Latest from Matthew Berman.
Matthew Berman · 16K views · 590 likes
Cerebras Killed Notion, Obsidian, and Your "Second Brain"
Latest from Nick Saraev.
Nick Saraev · 42.3K views · 1.5K likes
OpenAI gets sued for stealing, again…
Get $500 in Hyperagent credits to build agents that can ship real work - https://hyperagent.com/invite/FIRESHIP500 OpenAI is facing yet another lawsui
Fireship · 425.1K views · 13.2K likes

🗣 Voices & Blogs

Grok 4.5 vs. Claude Opus 4.8: Costs and what works, not the spec sheet
Can Grok 4.5 really match Opus for a quarter of the tokens? xAI released Grok 4.5 on July 8. It The post Grok 4.5 vs. Claude Opus 4.8: Costs and what… · The New Stack
GPT-5.6: Frontier intelligence that scales with your ambition
Agentic AI Summit Agenda Now Live and Tickets are Running Out + Livestream Sign-Up | Research Highlight: GPT-5.6's Agents' Last Exam Progress · via Berkeley RDI · Agentic AI Weekly

🌏 The Wire · Drudge / Breitbart

Helsing raises a $1.8 billion Series E at an $18 billion valuation, Europe's largest defense-AI round
The Munich defense-AI company, which builds software for autonomy and electronic warfare, closed the biggest round a European defense startup has raised, co-led by Lightspeed and General Catalyst. The signal: serious money is now flowing into AI for defense and sovereignty, not just consumer and enterprise apps.

🤖 Trending Models · Hugging Face

prism-ml/Ternary-Bonsai-27B-gguf
text-generation · ★827 · 338.9K dl
prism-ml/Bonsai-27B-gguf
text-generation · ★522 · 1.3M dl
baidu/Unlimited-OCR
image-text-to-text · ★2.3K · 2.1M dl

📈 Markets

NVDA 202.81 ▼2.2%
MSFT 393.82 ▼1.8%
GOOGL 346.77 ▼2.2%
AMZN 247.23 ▼1.1%
META 646.01 ▼2.8%
AMD 495.76 ▼1.0%
AVGO 370.82 ▼1.0%
PLTR 132.38 ▼1.5%
SPCX 123.99 ▼5.4%
TSLA 380.84 ▼2.6%

The BriefThe World AI Conference wrapped up in Shanghai this weekend, and the biggest news out of it is not a product. Twenty-nine countries signed the founding agreement for a new body called WAICO, the World AI Cooperation Organization, the first time governments have set up a standing intergovernmental group whose whole job is coordinating AI rules and safety. The initiative is led by China and headquartered in Shanghai. It matters because the fight over who writes the global rulebook for AI just moved from conference panels to an actual institution, and the standards it settles on will eventually shape which boxes your tools have to tick when you do business across borders.

Level UpGitHub just made Kimi K2.7 Code selectable inside Copilot, the first open-weight model in the picker and a genuinely low-cost option for everyday coding. If you use Copilot, the move this week is simple: open the model picker, switch one real task over to Kimi K2.7 Code, and run it side by side with whatever you default to. One afternoon of head-to-head use on your own work will teach you more about which model fits than any leaderboard. Try it: switch one task to Kimi K2.7 Code in Copilot