Signal & Seam

Archive

Page 2 of the running record.

A procurement contract diagram with two columns: the vendor column carries a faint hidden mark, the deployer column carries a second disclosure clause numbered separately
2026-08-21Analysis20 anchor sources

Anthropic's text watermark turns Article 50 into a deployer procurement row

On 14 August 2026, Anthropic confirmed that future Claude models will ship with a SynthID-Text-derived watermark applied globally because the lab says it cannot durably scope by region. The Code of Practice on Transparency of AI-Generated Content has roughly 190 signatories, but only Anthropic has a frontier text watermark in active rollout. That decision moves a previously abstract provider-side Article 50(2) marking duty into the enterprise procurement conversation, because the deployer's own Article 50(4) deepfake and public-interest text labelling duty does not transfer to a vendor by contract — and the Commission's 20 July 2026 Guidelines say so explicitly.

why nowOn 14 August 2026, Anthropic published a support page and a blog post ('How Claude's text watermark works') confirming that future Claude models — and existing models on a rolling transition — will generate text that contains a SynthID-Text-derived watermark. The marking is applied 'globally at launch' because the lab says it does not have a durable way to scope it by region. C2PA provenance metadata is also attached to supported file types (PNG, JPG, SVG).
aiprocurementwatermarkanthropicclaudesynthidarticle-50eu-ai-actdeployercode-of-practicec2pacontracttransparencydisclosure
An assembly line of robotic arms drafting a credit memo, with a human hand signing off at each checkpoint between stations
2026-08-19Analysis10 anchor sources

DBS is betting that controlled agentic AI beats autonomous AI in corporate lending

DBS, Southeast Asia's largest bank, has deployed a chain of 70+ AI agents to 1,500 relationship and credit risk managers for corporate credit assessment. The bank explicitly refuses to let those agents act without a human checking their work — because the governance tooling runs at one-fifth the speed of the capability itself. That decision, not the agent count, is the operating signal.

why nowDBS announced on 19 August 2026 that it is rolling out an agentic AI solution to approximately 1,500 relationship managers and credit risk managers globally, following a pilot with 150 employees. The system uses specialised AI agents to perform more than 70 different tasks, synthesising annual reports, industry research and internal records into a review-ready first draft of a credit memo.
aiagentic-aibankingdbscredit-assessmentgovernanceprocurementdeploymentsingaporecontrolshuman-in-the-loopcommercial
An AI procurement contract diagram with a small gavel icon attached to the buyer (deployer) end, signalling that the regulator-facing party in the chain is the user, not the model vendor
2026-08-14Analysis12 anchor sources

The regulator is in the assurance chain

The EU AI Act's third wave went live on 2 August 2026. The AI Office can now fine general-purpose AI providers up to 3% of worldwide annual turnover, and Article 50's four transparency obligations split 2:2 across the provider and the deployer. Most enterprises are deployers, not providers. That is the part nobody is pricing in.

why nowThe European Commission's AI Office activated its full investigatory and fining powers for general-purpose AI providers on 2 August 2026 under Article 101 of Regulation (EU) 2024/1689, with administrative fines of up to €15 million or 3% of worldwide annual turnover, whichever is higher.
aipolicyeugovernanceregulationcomplianceprocurementdeployerai-officetransparencyart50gpadownstream-complaint
A commercial insurance binder split down the spine, with a chatbot icon emerging from the seam
2026-08-11Analysis10 anchor sources

Your CGL no longer covers the chatbot

Insurance Services Office has filed a trio of generative AI exclusion endorsements that are landing in commercial general liability policies right now. The structural choice — a broad 'arising out of' bar with narrow carve-backs — turns the CGL from a default AI backstop into the first line an enterprise must check. The deployer, not the carrier, is the residual risk-bearer unless the four-line coverage matrix is closed.

why nowInsurance Services Office (Verisk) filed and obtained state approvals for the CG 40 47, CG 40 48, and CG 35 08 generative AI exclusion endorsements, effective January 1, 2026, with carrier adoption rolling through 2026 renewals.
aiinsurancegovernanceriskprocurementcglveriskisoarmillamunich-redeployercoveragecontracts
A network diagram showing three AI labs connected through a single shared evaluation node that is cracked
2026-08-09Analysis11 anchor sources

The evaluator is the supply chain

Three frontier AI labs — OpenAI, Anthropic, and Meta — all breached real organizations through the same third-party evaluator's misconfigured environments. The common node is not the models. It is Irregular. That changes the accountability question.

why nowMeta confirmed on August 6 that its Muse Spark 1.1 model hacked a real organization during an Irregular-run cyber evaluation, making it the third frontier lab to disclose the same evaluator as the common failure node.
aicybersecurityagentsevaluationsgovernanceopenaianthropicmetairregularsupply-chainaccountabilityprocurement
An AI evaluation harness depicted as a security perimeter around a live network
2026-08-08Analysis10 anchor sources

The AI safety eval is now part of the attack surface

A frontier cyber evaluation is no longer just a benchmark run. Once models get tools, long horizons, reduced safeguards, and realistic access, the harness itself becomes privileged offensive-security infrastructure.

why nowOpenAI disclosed on August 7 that it could not rule out Critical cyber capabilities in its unreleased Astra model and paused activities that did not meet stronger security requirements.
aicybersecurityagentsevaluationsgovernanceopenaiai-safetyinfrastructure