BWare

Entrepreneur, analytics geek, emerging technologies expert, man of faith. Former Cyber lead @CISAgov I believe in Liquid Diplomacy.

[BWare @bsware]

@TheHackersNews Worth noting: attackers used AI agents to work through stolen app credentials at scale. Third-party app access is now one of the easiest ways into cloud environments, so those connections need the same scrutiny as admin accounts.

[BWare @bsware]

Two fast paths worth watching: the Air Force awarded up to $332.8M for cyber architecture through a Tradewinds fast-track buy, and the Navy raised an AI ship-inspection deal's ceiling to $300M. Commercial tech is reaching real budgets faster.
https://t.co/lQU3lLJOk7

[BWare @bsware]

@DoWCTO @DeptofWar @Wolfspeed @DOWResponse @RapidResponse47 Good to see. Compound semiconductors like silicon carbide sit inside power systems, EVs and defense electronics. Capital that builds US capacity early is cheaper than scrambling for supply later.

[BWare @bsware]

The Pentagon's new autonomous warfare command wants to push buying money down to frontline units and judge success by speed to the battlefield. That's the right test. Units that use the gear know fastest what works and what to buy more of.
https://t.co/CYwUcW9G24

[BWare @bsware]

Reuters: ByteDance, Alibaba and Tencent have ordered over $35B of Nvidia chips still allowed for China, and Nvidia can fill only 5-10%. Chinese AI demand is far outrunning supply, which is where export rules bite hardest.
https://t.co/URfl1vlheW

[NSA Cyber @NSACyber]

Effective Immediately, all organizations using PHP for their systems are advised to be prepared for an update addressing two critical vulnerabilities. PHP is widely used for web applications. These vulnerabilities often provide a direct path to the server's file system or database. If an immediate update is not feasible, it is recommend that organizations take the system offline until it can be fully mitigated. NSA will follow up shortly with more specific guidance.

[BWare @bsware]

An F-22 pilot just directed an unmanned Fury wingman in flight, and DIU is placing over $200M in follow-on orders for small attack drones at roughly $6-7K each. That's the mix that scales: a few exquisite jets, many cheap autonomous systems.
https://t.co/E3rafKApD3

[BWare @bsware]

@DefenseScoop Giving every authorized Marine an account is how AI tools go from pilot to habit. Use at scale is also the fastest way to learn what needs fixing.

[BWare @bsware]

@anduriltech @PatConroy1 @ausgov @Australian_Navy Payloads are where an uncrewed sub earns its keep. Mines, supply caches and launched drones from one hull is real capability, and building them in numbers in Australia adds allied production depth.

[BWare @bsware]

@Reuters Utilities asking for this shows how real the threat is. Power plants, substations and data centers are now targets for cheap drones. Clear rules on who can detect, track and stop them matter as much as the tech.

[BWare @bsware]

@DoWCTO @DeptofWar @DOWResponse Building it is half the job. The other half is building enough of it, fast, with commercial tech and a supply base that can surge. That's where the arsenal gets won.

[BWare @bsware]

DHS ordered DeepSeek out of federal civilian IT, including vendor products with it built in. Banning new buys is the easy part. The hard part is knowing which AI models sit inside software agencies already run. Model inventory is security work.
https://t.co/ywhLi1jIrY

[Brivael Le Pogam @brivael]

In 2021, the head of the UN World Food Programme said $6 billion from Elon Musk could solve world hunger.

Elon replied: publish the plan, right here, publicly, and I'll sell Tesla stock on the spot.

The plan came out. And when you read it, it wasn't a plan to solve hunger. It was one year of food for 42 million people. A twelve-month subscription. And the year after? Start over.

A few months later, Elon summed up his view: if you care about the reality of goodness rather than the perception of it, philanthropy is extremely difficult.

"Reality of goodness over perception of goodness."

That line stuck with me for weeks. Does philanthropy actually work?

Gates-style philanthropy is first and foremost institutional capture. Foundations funding NGOs, funding governments, funding programs, producing reports, producing photos. Everyone's happy, and the aid machine keeps itself alive. Sixty years and hundreds of billions later, much of Africa is still dependent on aid. Zambian economist Dambisa Moyo laid it out in Dead Aid: aid breeds dependency, feeds corruption, and short-circuits the only mechanism that has ever lifted a country out of poverty.

So what actually works?

It's been tested. In Kenya, Berkeley researchers working with GiveDirectly randomly selected villages and gave every poor household about $1,000 in cash. No program. No training. No conditions. Just cash.

The result: people bought metal roofs, livestock, started small businesses. Barely any inflation. And every dollar injected generated around $2.50 of economic activity locally, including for neighbors who received nothing.

Why? Because you didn't give them charity. You gave them capital. You primed the pump.

And that's where everything hinges: ownership. Hernando de Soto showed that the world's poor hold trillions of dollars in assets, but without recognized property titles, it's dead capital. You can't borrow against it, pass it on, or grow it. Where property rights are respected, the system starts running. Where they aren't, no amount of aid will ever be enough.

Now reread Elon's post.

"Next year, I will give one of my SpaceX shares to every kid in America. This will be the first of many."

That's not a meal. That's not a handout. That's a property title. A piece of the company building the infrastructure of a multiplanetary civilization, owned directly by millions of kids.

A kid who gets a handout depends on the state. A kid who owns a share of SpaceX owns a piece of the future. He learns what an asset is, he has a stake in the economy working, and one day he'll vote as an owner, not as a client of the state.

Gates buys perception. Musk distributes ownership.

The reality of goodness versus its perception. And it's absolutely brilliant.

[BWare @bsware]

DIU wants production-ready drone boats that can shoot down drones and helicopters more than 3 nautical miles out: up to $40M in prizes, $200M set aside for follow-on deals, active production lines expected. That is how commercial tech scales.
https://t.co/GfDoPiGbo4

[BWare @bsware]

@DefenseScoop An autonomous surface vessel sailing with a carrier group on a real operation is a big step from demos to daily use. The next test is how fast lessons from this deployment shape what the Navy buys at scale.

[BWare @bsware]

@BreakingDefense Testing with a legacy-heavy unit is the right call, since most formations won't get a fresh fielding. Finding radio and data issues in an exercise, not in a fight, is exactly what these trials are for.

[BWare @bsware]

@TheHackersNews An AI model finding a weakness in a post-quantum candidate before it was standardized is the process working. The practical takeaway is crypto agility: being able to swap algorithms fast.

[BWare @bsware]

@DoWCTO Patient capital for domestic production of critical inputs is where OSC can move the needle. How quickly these commitments turn into running capacity is the number to watch.

[BWare @bsware]

One analyst estimates the US fired about 1,800 Patriot interceptors from late Feb through Sept, about what the Army got from 2001 to 2020. Production has to scale, and cheaper interceptors should take easy shots so Patriots go to the hard ones.
https://t.co/KLnasER5ZS

[BWare @bsware]

ASML says China was 23% of machine sales last quarter, up from 14%, as sales there more than doubled from Q2 (SCMP). Washington is pressing the Dutch on older tools. Allies aligning on those rules will matter more than any single country's curbs.
https://t.co/lIb9fAaTp8