Published
Five developer-relevant AI and technology shifts from this morning: governance crackdowns, energy pressure, and tooling updates that can change what teams ship next.
Tonight’s rundown
Financial Times-sourced reporting says Anthropic is moving to close loopholes that let Chinese firms route usage past restrictions. The report says firms like Ant were using workarounds through cloud providers and overseas subsidiaries. Engineers are still finding ways to use the models despite the tighter controls.
Techmeme · AI
A Techmeme summary reports that Anthropic is taking steps to close access loopholes.
The report says Chinese firms like Ant had been using workarounds to keep using models despite restrictions.
Those routes are described as using cloud providers and overseas subsidiaries.
The same item says engineers are still finding ways to use the models even with tighter controls.
close loopholes that let Chinese firms like Ant use its models via workarounds including cloud providers and overseas subsidiaries
A Hacker News repost reports that Alibaba has banned employees from using Claude Code in the workplace and asked staff to remove all Claude models from their work computers. The move is tied to alleged backdoor and security risks. For developers, it is a clear example of enterprise AI tooling risk becoming immediate operational policy.
Hacker News · SECURITY
The post links to a Reuters item about an internal rollout at Alibaba to ban Claude Code usage.
The report says employees were told to remove all Claude models from work computers.
The stated reason is alleged security risk from possible backdoor behavior.
The decision is company-specific but the pattern is general: AI tooling can be removed quickly when risk signals rise.
Security decisions on coding assistants are now governance decisions, not product preferences.
Ars Technica reports that Google's AI buildout caused a 37% increase in electricity use in 2025. The story reframes AI growth as an infrastructure and operations concern, not just a product feature race. Teams shipping AI systems now face tighter ceilings on power budgets and thermal planning.
Ars Technica · BIG TECH
The company’s latest sustainability report acknowledges that Google’s total electricity usage has increased by more than 250 percent since 2019, which the company attributed to ongoing growth in Google Cloud, YouTube video streaming, and data center construction and operations supporting various AI products and services.
“While the path to achieving our climate ambitions will not be linear—given our AI infrastructure buildout is currently accelerating faster than the grid is decarbonizing—we remain focused on scaling abundant and affordable clean power globally and progressing technological innovations that drive down emissions across our operations and the broader industry,” according to the Google sustainability report.
As the bulk of electricity usage, Google’s data centers consumed more than 42 million megawatt-hours of electricity in 2025 compared to 30.6 million megawatt-hours of electricity in 2024. That means Google’s data center energy usage rivals the electricity consumption of entire countries such as New Zealand, Denmark, and Nigeria.
Despite the steep rise in electricity consumption, Google reported actually reducing its operational emissions by 2 percent over the same year period. Such apparent “decoupling of electricity-related emissions” from growth in electricity usage is potentially promising, but Google said it would need to boost “clean energy investments and closer partnerships with local stakeholders in the years ahead.”
But the company also noted that its supply chain emissions from its contracted manufacturers and suppliers actually grew by 25 percent because of “an Asia-Pacific supply chain operating on grids that remain undersupplied with carbon-free energy.” As a result, Google reported total “ambition-based emissions” increasing by 18 percent between 2024 and 2025.
AI buildout drove 37% increase in electricity use in 2025
A Techmeme report relaying Bloomberg says SEMI warned that policies affecting chip prices or production capacity would worsen the shortage. The letter names major firms including Micron and Samsung, emphasizing that shortages are tied to policy and market timing. For AI infrastructure, this matters because memory and compute supply risk now flows through regulatory decisions.
Techmeme · CHIPS
The techmeme item reports a Bloomberg source that SEMI sent a formal warning to U.S. policymakers.
The warning says actions affecting prices or production capacity could worsen the existing chip shortage.
It points to major industry participants such as Micron and Samsung in support of the position.
The practical read is that semiconductor scarcity can become a direct planning constraint for AI and cloud-heavy roadmaps.
A Techmeme report relaying Bloomberg says SEMI warned that policies affecting chip prices or production capacity would worsen the shortage.
A Hacker News post links to WebKit’s announcement of a Safari MCP server for web developers. The change adds another browser option for MCP-based tooling workflows. It is directly relevant for teams that are standardizing model-context and browser automation stacks.
Hacker News · DEV
WebKit announced a Safari MCP server aimed at web developers.
The post positions it as a developer-side tool addition in the browser stack.
No full technical changelog details are available in the scrape beyond the launch itself.
For teams, this is a practical signal to evaluate Safari in MCP-centric automation and testing pipelines.
Browser tooling is becoming the new integration layer for AI workflows.