- How t54 built a trust layer with Amazon Bedrock AgentCore paymentsFactual summary
t54 deployed x402-secure on Amazon Bedrock AgentCore to govern over 20 million autonomous agent payments with session budgets.
- EMVCo reqests feedback on framework for card-based agentic paymentsFactual summary
EMVCo released a draft framework establishing security and interoperability standards for card-based agentic payment execution.
- Big banks want a cut of lawyers’ AI savingsFactual summary
Goldman Sachs, Morgan Stanley, and Citi are pressing external law firms to lower fees, citing productivity gains from legal AI tools.
So whatPeer banks are shifting vendor management strategy to capture third-party AI efficiency gains in external legal spend.
Do whatBrief your legal operations procurement team to factor external counsel AI adoption into contract renewal negotiations.
- Autohealing MoneybotFactual summary
Cash App built a system to reproduce stochastic LLM failures, verify fixes statistically, and automate pull requests.
- OpenAI’s Astra model is on the way — and very good at breaking into computer systemsFactual summary
OpenAI previewed safety precautions for Astra, an upcoming LLM with advanced cyber-offensive capabilities for breaking into systems.
- Retail Investing Opens Up to AI AssistantsFactual summary
Scalable Capital opened its investment platform to external AI assistants, allowing direct integration with user brokerage accounts.
- OpenAI Is About to Release Its First AI Model With ‘Critical’ Cyber AbilitiesFactual summary
OpenAI will provide select partners early access to its Astra model to evaluate its advanced offensive and defensive cyber capabilities.
- Building a transaction foundation model to power intelligent finance | PlaidFactual summary
Plaid has introduced a proprietary transaction foundation model aimed at improving financial data categorization and contextual insights.
- Connect Amazon Bedrock AgentCore to cross-account knowledge basesFactual summary
AWS details a cross-account Amazon Bedrock AgentCore architecture to query Bedrock knowledge bases and Redshift without data replication.
- On Benchmark Hacking in ML Contests: Modeling, Insights and DesignFactual summary
Research paper models benchmark hacking in ML contests, showing how models are tuned to score highly without true generalization.
So whatThis research provides a framework for understanding and mitigating benchmark hacking, which directly impacts the reliability of internal model validation and external vendor evaluations.
Do whatYour model validation team needs to integrate considerations of benchmark hacking into evaluation protocols for both in-house and third-party models.