Eyes on the Chaos
Sunday, May 31, 2026

Archived edition

Sunday, May 31, 2026

6 stories curated from 16 sources

In today's issue

DesignEthicsProduct
  1. 01
    AI grifters are creating fake Black people to sell Shein junk

    Fake AI-generated Black influencers are selling cheap products on TikTok.

  2. 02
    SoftBank says it will invest up to €75 billion to build French data centers

    SoftBank commits massive investment for 5 gigawatts of French data capacity.

  3. 03
    I put Google's 24/7 AI assistant Gemini Spark to work, and it's actually pretty useful

    Google's Gemini Spark automates daily tasks effectively but puzzling product strategy.

  4. 04
    'What a joke': Github Copilot's new token-based billing spurs consternation among devs

    GitHub Copilot switches to usage-based billing, angering developers with unpredictable costs.

  5. 05
    Powerful A.I. Super PACs Duel Over the Midterms: 'This Is a War'

    OpenAI and Anthropic-linked super PACs spend millions influencing midterm elections.

  6. 06
    How one founder's bet on 'the old school web' is paying off

    Former Meta engineer rejects AI funding to build profitable old-school website.

AI Research & News

AI grifters are creating fake Black people to sell Shein junk

The Verge

EthicsProduct

Fake AI-generated Black influencers are selling cheap products on TikTok.

  • The deception: AI-generated personas like 'Aliyah' appear as crying Black women pleading for support for fake handmade businesses. They're actually selling mass-produced items from platforms like Shein.
  • Exploitation angle: Scammers specifically target racial identity and sympathy to manipulate viewers into purchasing products. The fake personas use emotional appeals about supporting Black-owned businesses.
  • Platform blindness: TikTok's content moderation appears unable to catch these sophisticated AI-generated seller accounts. The fake influencers are proliferating across the platform.

For product

Review your platform's AI detection capabilities for user-generated content — this attack vector could easily spread to other social commerce features.

SoftBank says it will invest up to €75 billion to build French data centers

TechCrunch

SoftBank commits massive investment for 5 gigawatts of French data capacity.

  • Scale commitment: SoftBank plans to invest up to €75 billion in French data center infrastructure. The goal is developing 5 gigawatts of additional capacity.
  • AI infrastructure: This massive investment reflects the enormous infrastructure requirements for AI model training and deployment. Europe is becoming a key battleground for AI compute capacity.
  • Geopolitical play: The investment positions SoftBank and France strategically in the global AI infrastructure race. Other regions are likely planning similar massive investments.

Product & UX

I put Google's 24/7 AI assistant Gemini Spark to work, and it's actually pretty useful

TechCrunch

ProductDesign

Google's Gemini Spark automates daily tasks effectively but puzzling product strategy.

  • What it does: Gemini Spark handles routine automation like inbox summaries, event planning, and task management. It runs continuously in the background unlike traditional chat interfaces.
  • User experience: The reviewer found it genuinely useful for everyday productivity tasks. It works more like a proactive assistant than a reactive chatbot.
  • Product confusion: It's unclear why Google launched this as a separate product rather than integrating it into existing Gemini. The fragmented approach may confuse users about which Google AI tool to use.

For product

Consider whether your AI features need separate apps or should integrate into existing workflows — user adoption suffers when people can't figure out which tool to use for what.

Business & Strategy

'What a joke': Github Copilot's new token-based billing spurs consternation among devs

TechCrunch

Product

GitHub Copilot switches to usage-based billing, angering developers with unpredictable costs.

  • Pricing shift: GitHub moved from flat monthly fees to token-based billing where costs fluctuate based on actual AI usage. Developers now face unpredictable monthly bills.
  • Developer backlash: Users are calling the change a 'joke' and expressing frustration over losing cost predictability. Many are threatening to switch to competing AI coding tools.
  • Market signal: This reflects broader pressure on AI companies to move from subsidized growth pricing to sustainable revenue models. The 'golden age' of cheap AI tools may be ending.

For product

If you're planning AI feature rollouts, lock in pricing models early and communicate changes transparently — developer trust is fragile and competitors are ready to capitalize.

Powerful A.I. Super PACs Duel Over the Midterms: 'This Is a War'

NYT Technology

Ethics

OpenAI and Anthropic-linked super PACs spend millions influencing midterm elections.

  • Corporate politics: Major AI companies are now directly funding political campaigns through affiliated super PACs. One is tied to OpenAI, the other to Anthropic.
  • Regulatory stakes: The companies are spending millions to influence elections that will shape AI regulation and policy. They're treating political influence as essential business strategy.
  • Industry warfare: Sources describe the political competition between AI companies as 'a war.' The stakes are high enough that companies are willing to fund opposing political candidates.

For ethics

Expect increased scrutiny of your company's AI partnerships and vendor relationships as political affiliations become part of procurement decisions.

How one founder's bet on 'the old school web' is paying off

The Verge

Product

Former Meta engineer rejects AI funding to build profitable old-school website.

  • Counter-narrative: Craig Campbell walked away from AI investor money to create Past Maps, a simple website selling historical maps. VCs pressured him to start an AI company instead.
  • Profitable simplicity: The 'old school web' approach is apparently generating sustainable revenue without venture funding. The business model relies on straightforward e-commerce rather than AI complexity.
  • Market timing: While everyone chases AI opportunities, there may be underserved markets for simple, well-executed web experiences. The founder bet against the prevailing investment wisdom.

For product

Don't assume every product needs AI features — sometimes the biggest opportunity is solving problems with simpler, more reliable approaches while competitors overcomplicate.