A library of snippets
xxx
But a growing number of forecasters now say the economy’s dependence on AI was overstated. Prominent economists, including from Morgan Stanley and JPMorgan Chase, calculate that the AI buildup was directly responsible not for 92 percent or 39 percent of gains to the U.S. economy in 2025, but as little as zero.
From: How much did AI boost the economy? Maybe zilch, some economists say. – The Washington Post.
xxx
xxx
Even if Moltbook does not spell the imminent subjugation of humanity, it poses other risks. Some careless users are running up thousands of dollars in cloud-computing fees as their agents draw on cutting-edge ai models to function. Then there are the scammers, who are taking advantage of the free rein that OpenClaw agents have over the devices on which they run on. Already Moltbook has been inundated by attempts (including by humans pretending to be bots) to convince ai agents to hand over cryptocurrency. The strange experiment could well prove costly—and short-lived
From: A social network for AI agents is full of introspection—and threats.
xxx
xxx
merchants increasingly treat unidentified automation as a policy and risk problem. Amazon’s lawsuit against Perplexity over agentic shopping makes the direction clear.
– The path forward is not “smarter bots clicking websites.” It’s agent-native commerce interfaces: OpenAI and Stripe’s ACP, Google’s UCP, and browser-level standards like WebMCP.
xxx
xxx
There are 1.5 million agents transacting on Moltbook right now. Depending on who you ask, this is either the early singularity, a dumpster fire, or 17,000 humans puppeting bots. Wiz Research found 341 malicious skills on ClawHub stealing credentials. Karpathy called it what it is.
From: What Agents Need Before They Handle Real Money – Catena Labs.
xxx
xxx
Most agent frameworks treat policy as an application-layer concern. The AI decides whether a transaction should happen, checks some rules in code, hopefully works within prescribed guardrails, and proceeds. This is fine for demos. It is not fine for real money.
The problem is straightforward: application-layer policy is only as secure as the application. If someone compromises the server, jailbreaks the model, or finds a bug in your policy-checking code or guardrails framework, the money moves. You’ve built a lock out of suggestions.
What agents actually need are two layers:
Layer 1 is intelligence. This is the application layer — the part that answers the questions you’d want answered before any money moves. Who controls this agent? Are they a verified entity? What’s their track record? You can see this in the demo: before the treasury agent pays another agent for research services, it resolves their identity, checks their reputation score, and evaluates whether they meet the policy threshold. An agent with a verified owner, a score of 87/100, and 142 attestations clears. An unverified agent with a dispute flag doesn’t. This is the kind of automated standards-based trust infrastructure that the agentic economy needs — not platform-specific API keys, but portable, verifiable identity that works across any agent framework.
Layer 2 is enforcement. In this example, this part runs in Turnkey’s secure enclave. Has the required approval been obtained? The enclave signs the transaction only if every policy condition is met. This isn’t running in our application code. It’s running in hardware that neither we nor Turnkey can tamper with after deployment.
Intelligence without enforcement is just prompt suggestions. Enforcement without intelligence is just a dumb access list. You need both.
Even if our entire backend is compromised, the enclave won’t sign transactions that violate policy. That’s not a promise — it’s math.
From: What Agents Need Before They Handle Real Money – Catena Labs.
xxx
xxx
France’s Ministry of Economics, Finance and Industrial and Digital Sovereignty last week revealed the incident took place in January, after unknown attackers used stolen credentials to access the database.
The Ministry said the attacker’s access was restricted immediately upon discovery of the attack, but that the miscreant still managed to access personal information about 1.2 million accounts, including account numbers, account holder’s addresses, and tax identification numbers.
From: Attacker gets into France’s DB listing all bank accounts • The Register.
These purloined personal parameters will undoubtedly be used for social engineering attacks against account holders.
xxx
police launched an appeal after someone sold a bike for £4,700 in fake cash.
xxx
xxx
On a larger scale, the entire agentic world is wrestling with that trade-off: At what point will agents be secure enough to be useful?
From: Is a secure AI assistant possible? | MIT Technology Review.
xxx
xxx
Dark Reading reported this week on Operation DoppelBrand, a campaign by the GS7 cyberthreat group deploying near-perfect imitations of U.S. financial institution login portals. These aren’t phishing emails with typos. These are pixel-perfect clones of major bank portals hosted on lookalike domains, designed to steal credentials and establish remote access.
Most brand protection tools focus on domain monitoring—watching for typosquatting and lookalike URLs. But these attacks exploit user trust in visual design, not just domain names. And let’s face it, who REALLY checks the URL when they go to their bank’s website? Me neither.
From: This Week in Fraud (2/17).
xxx
xxx
Figure Technology, a $5 billion blockchain-based lender that tokenizes home equity loans on its own distributed ledger, just learned the hard way that you can have the most secure vault in the world, but that’s pointless if your employees give out the keys.
On February 13th, Figure confirmed a data breach after an employee fell victim to a voice phishing (vishing) attack. The attacker impersonated IT support, tricked the employee into surrendering their Okta single sign-on credentials, and used a real-time Adversary-in-the-Middle (AiTM) phishing kit to bypass multi-factor authentication entirely. ShinyHunters, one of the most prolific ransomware groups operating today, published roughly 2.5 gigabytes of stolen data after Figure refused to pay the ransom demand.From: This Week in Fraud (2/17).
xxx
Snippings