Alibaba rolls out new AI chip

Alibaba builds its own path for AI chips

The chip is called C950.

Alibaba just introduced a new CPU through its Damo Academy, built on RISC-V and aimed at cloud inference workloads where companies want more control over how models run.

It matters because Alibaba is trying to reduce reliance on external chip suppliers while still keeping up with rising demand for AI workloads that are getting more specialized and cost sensitive by the quarter. You can hear the hum of server racks in this move, since inference is where real usage happens and where margins get decided.

This looks practical. It gives customers flexibility to tune chips for specific tasks, which could lower costs if it works as promised.

If more Chinese firms lean into RISC-V designs like this C950, the global chip stack starts to fragment in a way that favors regional ecosystems and local control.

Will developers actually switch?

Apple signals it’s ready to rejoin AI race

Apple is finally talking about AI again.

At WWDC on June 8, the company plans to highlight “AI advancements,” with iOS 27 expected to bring a more capable Siri and possibly a chatbot-style version.

The details are still thin, but the direction is clear: Siri gets deeper context, can see what’s on your screen, and works across apps in a way it hasn’t before. This is the upgrade Apple has been delaying since iOS 18, and expectations are already set by tools like ChatGPT and Gemini.

This feels late. Apple waited while others trained users to expect fast, flexible AI assistants, and now it has to meet a bar it didn’t define.

Picture someone tapping their phone at night, asking Siri something it used to fumble, and waiting to see if this time is different.

If Apple gets this right, it tightens its ecosystem again. If it doesn’t, users will keep reaching for other apps.

Will people give Siri another shot?

Security tools race to track AI agents

This is about control.

Palo Alto Networks just launched Prisma AIRS 3.0, a platform designed to track, assess, and secure AI agents across their full lifecycle as companies move toward autonomous systems.

The pitch is simple: companies can see what their AI agents are doing, not just what they say, and apply controls as those agents take actions across cloud, SaaS, and endpoints. It also pulls scattered tools into one system, which matters when teams are already juggling too many dashboards and partial views.

This feels like a necessary catch-up. The shift to agentic AI creates real exposure, and most companies are not set up to monitor software that acts on its own in production systems.

You can picture a late-night operations floor with screens flickering, and no clear answer on which agent triggered what action or touched which data.

Palo Alto says it serves 70,000+ customers, and this looks like an attempt to lock them into its view of how AI security should work.

If agents become normal, whoever owns visibility wins.

Who actually gets that visibility first?

Prohuman team

Free Guides

Explore our free guides and products to get into AI and master it.

All of them are free to access and would stay free for you.

Feeling generous?

You know someone who loves breakthroughs as much as you do.

Share The Prohuman it’s how smart people stay one update ahead.