SunBrief#86: How AI Helped Treat Cancer

Welcome to the SunBrief

Today in SunBrief 🌞

• 1,000+ ChatGPT Prompts from Superhuman AI

• Founder Battles Rare Cancer Using AI as a Medical Copilot

• Stock Updates

• TikTok is quietly becoming a super app

• Anthropic Accuses Alibaba of Largest Claude Cloning Attempt Yet

• AI Highlights of the Week

• Too Important to Miss

Supercharge Your Productivity With 1,000+ ChatGPT Prompts from Superhuman AI

ChatGPT is insanely powerful.
But most people waste 90% of its potential by using it like Google.
Master AI and work 10X faster with 1,000+ proven ChatGPT prompts.
Sign up for Superhuman AI to get:

• 1,000+ ready-to-use prompts to solve problems in minutes instead of hours—tested & used by 1M+ professionals

• Superhuman AI newsletter (3 min daily) so you keep learning new AI tools & tutorials to stay ahead in your career—the prompts are just the beginning

Founder Battles Rare Cancer Using AI as a Medical Copilot

Conno Christou turns to wearables, biomarker data, and Claude to navigate diagnosis, treatment, and recovery decisions

Tech founder Conno Christou, known for extreme health optimization, was diagnosed with aggressive non-Hodgkin's lymphoma after a post-workout issue. It was discovered during pre-surgery scans for blood clots and found to be fast-growing and life-threatening.

Key Points:

• Unexpected diagnosis: Despite years of perfect health tracking and clean annual biomarker reports, doctors discovered an 11×11×8 cm tumor behind his sternum.

• Fast progression: The cancer was estimated to have developed over just three months, with the potential to reach stage four within weeks.

• Data-driven treatment choice: Christou gathered 12 expert opinions and chose the more aggressive regimen based on consensus from specialists.

• AI-assisted health tracking: During treatment, he fed bloodwork, scans, wearable data, and journal entries into Claude, using it to detect patterns and guide questions for doctors.

• Wearables as early signals: Devices like Whoop and Oura helped him anticipate immune system crashes before symptoms appeared.

• AI helped avoid overtreatment: Near the end of therapy, Claude flagged a likely thymus rebound effect—a known false-positive in post-chemo PET scans, helping avoid unnecessary radiotherapy.

Why It Matters:
This case shows AI becoming a medical decision support layer, not replacing doctors but helping interpret complex data. It also highlights a shift where wearables, lab results, and AI combine into continuous health intelligence, especially useful in rare or unclear conditions where experts may disagree.

Stock Updates

TikTok is quietly becoming a super app

Short-form video platform evolves into an all-in-one digital ecosystem competing with Google, Amazon, and fintech apps

TikTok is evolving into a super app, adding shopping, search, travel booking, payments, gaming, and scripted entertainment, similar to China’s WeChat.

Key Points:

• Super app strategy: TikTok is expanding beyond video into a full ecosystem for commerce, payments, entertainment, and services inside a single app.

• TikTok Shop growth: Its in-app marketplace is a major success, reaching an estimated $15.8B in U.S. sales, making TikTok a serious competitor to Amazon and Shein.

• Search & maps: TikTok now offers location discovery, reviews, and map-style search results, reducing reliance on Google Search and Maps.

• Travel bookings: “TikTok GO” allows users to book hotels and experiences directly from videos, turning content into transactions.

• Fintech push: The company has applied for licenses to offer payments, stored balances, and lending services, signaling deeper fintech ambitions.

• Entertainment expansion: TikTok is experimenting with microdramas, casual games, and extended content formats to increase engagement.

• Music experiment failed but influence remains: TikTok shut down its standalone music app but still integrates with Apple Music to keep users in-app.

Why It Matters:
TikTok is moving beyond social media into a multi-industry platform, overlapping with Google in search, Amazon in commerce, Netflix in entertainment, and fintech apps. If successful, it could become a global super app outside China, changing how users search, shop, pay, and consume content in one place.

Anthropic Accuses Alibaba of the Largest Claude Cloning Campaign Yet

The company says Alibaba-linked operators used 25,000 fake accounts to mine Claude’s advanced capabilities

Anthropic has accused Alibaba-linked operators of running the largest known effort to extract Claude’s capabilities, allegedly creating 28.8 million exchanges through nearly 25,000 fake accounts to replicate advanced model behavior.

Key Points:

• Massive alleged extraction: Anthropic says the campaign ran from April 22 to June 5 and targeted Claude’s agentic reasoning, coding, and long-horizon task abilities.

• Alibaba accused: Anthropic claims operators tied to Alibaba and Alibaba Qwen used fake accounts, proxy networks, and obfuscation techniques to avoid detection.

• Distillation concern: Anthropic argues the goal was to copy Claude’s capabilities without paying the enormous training and R&D costs needed to build a frontier model.

• Congress alerted: Anthropic shared the claims in a confidential letter to Senators Tim Scott and Elizabeth Warren ahead of a Senate hearing on AI.

• Punishment requested: Anthropic wants stronger U.S. penalties for foreign labs that illicitly extract model capabilities, including limits on access to U.S. models, chips, and data centers.

• China competition angle: Anthropic warned that these attacks could help Chinese labs close the gap with advanced U.S. models like Mythos Preview.

• Alibaba under pressure: The accusation comes as Alibaba is already challenging U.S. government restrictions and denies links to the Chinese military.

Why It Matters:
This shows frontier AI is shifting beyond model development into security, espionage, and geopolitics. If models can be replicated through large-scale usage, labs may tighten access and push stronger regulation as AI becomes critical to cybersecurity and defense.

AI Highlights of the Week

• OpenAI Previews GPT-5.6 With Stronger Cybersecurity AI

  OpenAI unveiled a limited preview of GPT-5.6, introducing three new models with improved reasoning, lower costs, and stronger cybersecurity capabilities.

  The release adds enhanced jailbreak protections and prepares the models for a wider rollout in the coming weeks.

• Apple Vision Pro Executive Joins OpenAI
  Apple executive Paul Meade, who led Vision Pro engineering, is reportedly leaving to join OpenAI's hardware team.

  The move strengthens OpenAI’s AI device ambitions as it works with Jony Ive on a next-generation AI product.

• Anthropic Restores Limited Access to Mythos 5
  Anthropic regained limited access to Mythos 5 after the U.S. government partially lifted its two-week restriction.

  The move restores the model for approved U.S. organizations, while Fable 5 remains suspended amid ongoing security reviews.

• China’s Zhipu Challenges Top U.S. AI Models
  Zhipu’s GLM 5.2 is approaching the performance of Anthropic and OpenAI models while costing a fraction as much.

  The open-source model offers lower costs and enterprise-ready AI, gaining momentum as U.S. frontier models face government restrictions.

Too Important to Miss

Newegg wants AI to build your next gaming PC

Qualcomm plans new chip architecture for phones

Scientists built an AI that can design healthier burgers

Meta Brings Back Facebook Creator Studio With AI Features

Last Week’s Poll Result

• Is Anthropic becoming a real threat to Google DeepMind in scientific AI?

  Maybe, but DeepMind still leads → 41.67%

  Yes, this move proves it → 33.33%

  No, one hire doesn't change the field → 25.00%

  

• Should Amazon sell Trainium chips directly to third-party data centers?

  Maybe, if supply can keep up → 66.67%

  No, AWS should keep them inside its cloud → 33.33%

  Yes, it opens a huge new market → 0.00%

• Should frontier AI companies be treated as national security risks?

  Yes, advanced AI is strategic technology → 44.44%

  Maybe, depending on the model → 33.33%

  No, this goes too far → 22.22%

Feedback

Know someone who may be interested?