InsightsLatest
Latest from Drip
Deep dives, recaps, weekly briefs, trends, and expert answers, newest first.
Trend
Australia’s Mine Surge Collides with a Conversion Shortfall
Australia’s lithium sector is adding mine supply faster than it can build domestic conversion, creating a structural gap between extraction and battery-grade output. That leaves EV supply chains dependent on offshore processors even as policy pressure increases the premium on traceable, low-carbon materials. The next competitive chokepoint is shifting from feedstock control to processing access, with conversion, recycling, and magnet capacity becoming the most durable sources of supply security.
Trend
SK On and Nano One Push LFP Into Process Qualification
SK On and Nano One show that the LFP race is shifting from lab performance to industrial readiness. By simplifying production steps and advancing material sampling into the customer-evaluation stage, they are helping make cheaper chemistries easier to qualify and scale. That matters because EV pricing is now being set by how quickly suppliers and OEMs can industrialize LFP at volume, not just by chemistry selection.
Trend
Frax, Kraken, and CME Push Regulated Access to the Front End
Regulated access is becoming the new battleground in DeFi as firms like Frax, Kraken, and CME bring crypto exposure into compliant banking, brokerage, and derivatives channels. That shift matters because it lowers friction for institutions and mainstream users while moving control toward the firms that own custody, compliance, and market access. The next phase of DeFi growth is likely to come less from permissionless usage and more from regulated wrappers that make on-chain assets tradable inside existing financial systems.
Trend
Bon Secours and RadNet Buy the Operating Layer Beneath Volume
Healthtech consolidation is moving deeper into the operating layer beneath patient volume, with buyers acquiring the sites, labor, and infrastructure that make care delivery run. Deals like Bon Secours’ Surgery Partners buyout and RadNet’s ONRAD acquisition show that control of outpatient capacity, radiology coverage, and recurring service contracts is becoming more valuable than standalone software or point solutions. The market is heading toward fewer platforms that own both workflow and the operational assets behind it.
Trend
India Flags Agent Spillover as Microsoft Extends Sovereign AI Controls
India’s CERT-In advisory and Microsoft’s sovereign cloud expansion point to the same market shift: agentic AI now needs containment, not just capability. As multi-agent systems spread, security and compliance requirements are moving into the product layer, with residency, audit trails, and policy enforcement becoming core buying criteria. The next phase of enterprise AI will favor vendors that can prove control over data, actions, and infrastructure.
Trend
Salesforce Starts Billing for Headless Platform Interactions
Salesforce’s new Headless Platform Interactions policy makes AI-agent execution inside its platform a monetized event, not just a technical integration. That matters because it extends the industry’s pricing shift from seats to usage and toward outcome-linked economics, while forcing buyers to register agents, track consumption, and manage margin more tightly. The next phase of SalesTech monetization is likely to center on authenticated agents, metered actions, and auditable business impact.
Trend
Salesforce Brings Execution Offline
Salesforce is pushing beyond cloud-connected workflow automation into an offline-first execution layer for field operations. By acquiring Rune Technologies and its TyrOS system, Salesforce can support local AI decisioning, low-connectivity device workflows, and later synchronization for mission-critical environments. That matters because it expands the platform from managing enterprise processes to executing them where networks are unreliable, raising the bar for resilient, domain-specific AI.
Trend
Attackers Start Monetizing the AI Stack Itself
Attackers are monetizing the AI stack itself by turning cloud accounts, model endpoints, and billing systems into attack infrastructure and cost-extraction targets. That shift makes AI a direct operational risk, not just an enabler of faster phishing or reconnaissance, because it lets adversaries scale intrusion while also imposing financial damage on victims. The trend points toward more abuse of legitimate AI services, more automated attack workflows, and stronger demand for controls across identity, cloud, and AI governance layers.
Trend
ServiceNow, Microsoft, and Spacelift Extend the Execution Layer
ServiceNow, Microsoft, and Spacelift are extending the execution layer by embedding AI into systems that can act on work, not just recommend it. That shift is pushing operations from static planning toward continuous orchestration, where staffing, workflows, and exceptions are managed in real time across human and AI labor. The next phase is tighter closed-loop control, with faster response times and stronger guardrails around accountability.
Trend
Saudi Funding Slows as Follow-On Capital Gets Harder to Reach
Saudi startup funding has shifted from a pricing problem to an access problem, with later-stage and follow-on capital becoming much harder to reach. The slowdown is most visible in the collapse of pre-seed and seed activity and the absence of Series B and C+ rounds, even as Series A remains comparatively resilient. For founders, the market is moving toward earlier fundraising, smaller rounds, and stricter milestone credibility requirements.
Trend
PR Measurement Vendors Turn AI Visibility Into a Managed Workflow
PR measurement is evolving from periodic reporting into an always-on AI visibility workflow. Vendors now package prompt testing, cited-source analysis, recommendation tracking, and attribution so teams can monitor how often brands appear and how they are framed across major AI systems. The next phase is operational discipline: using these tools to improve retrievability, citation quality, and recommendation frequency over time.
Trend
Conga and Workday Turn Contract Data into Operational Triggers
Conga and Workday are pushing contract management beyond monitoring into execution, using contract data to trigger downstream actions in ERP, procurement, billing, and HR workflows. That shift matters because legal terms are becoming operational controls, not just records, which raises the stakes for governance, exception handling, and auditability. The next phase of CLM is likely to center on agentic orchestration with legal teams supervising rules and thresholds rather than manually chasing deadlines.
Trend
Red Hat, Nokia, and Extreme Turn Sovereign AI Into Deployable Infrastructure
Red Hat, Nokia, and Extreme are showing that sovereign AI is becoming an infrastructure problem as much as a governance one. Their platforms emphasize repeatable provisioning, isolation, and operational control so regulated organizations can deploy AI without losing residency or security guarantees. The direction of travel is toward standardized, automatable sovereign stacks that make compliance easier to enforce at scale.
Trend
PagerDuty, Auvik, and ConnectWise Put Agents Inside the Incident Workflow
AI agents are moving into the incident workflow itself, where they can detect issues, investigate root causes, and carry out low-risk remediation under policy controls. PagerDuty, Auvik, and ConnectWise show the same pattern: embed the agent where operators already work, then use approvals, allowlists, and rollback logic to keep automation safe. The next competitive edge will come from teams that can turn troubleshooting knowledge into governed, repeatable remediation playbooks.
Trend
Apple and Meta Turn AI Capacity Into a Managed Portfolio
Apple and Meta are showing that AI capacity is no longer just a scaling problem; it is a portfolio management problem. By pooling idle devices, reserving GPU headroom for recovery, and routing requests to lower-cost models or infrastructure, IT teams can reduce spend without sacrificing throughput or reliability. The trend points toward more centralized control of AI workloads, with scheduling, chargeback, and observability becoming essential to unit economics.
Trend
symplr and Dayforce Add Agentic Capacity Orchestration to Workforce Planning
Workforce planning is moving from dashboards and forecasts to agentic decision support that can recommend staffing interventions, compare scenarios, and surface cost and skills tradeoffs. Vendors like symplr and Dayforce are extending planning into live capacity orchestration, while broader HCM platforms are adding AI guidance for hiring, redeployment, and org design. The direction is clear: People Operations teams will need to govern AI-driven recommendations, not just interpret reports.
Trend
GSK’s Agentic Pharmacovigilance Test Pushes QA Into Live Evidence Chains
Agentic pharmacovigilance is pushing QA/QC toward live evidence chains: AI agents can execute case intake, coding, checks, and signal work, but every step must remain traceable, reviewable, and governed. That changes QA from periodic validation to continuous proof of control, with emphasis on provenance, audit trails, human approval, and record integrity. The direction of travel is clear: regulated workflows will increasingly be judged by how well they generate defensible evidence in real time.
Trend
Keyhaven’s Novation Tech CV Shows the Market’s Next Step
Keyhaven’s Novation Tech CV signals that continuation vehicles have entered a more mature phase: the market is now testing how to extend ownership of high-performing assets while preserving a credible liquidity option for LPs. That matters because the focus is shifting from whether CVs belong in the toolkit to how selectively and transparently they are deployed. For PE and growth investors, the next step is more disciplined execution around valuation, elections, conflicts, and hold-extension underwriting.
Trend
Hamilton Lane, FactSet, Solovis, and PitchBook Turn Portfolio Monitoring Into Live Query Workflows
Portfolio monitoring is moving from periodic reporting to live, query-driven workflows across private equity and growth investing. Vendors are embedding AI into monitoring, benchmarking, risk analysis, KPI extraction, and alerting so teams can interrogate portfolio data in real time rather than assemble it manually. The next phase is a continuous surveillance layer that standardizes data, surfaces exceptions, and pushes analysts toward higher-value judgment and intervention.
Trend
Stripe, Microsoft, and BlackLine Put Agent Permissions Into Production
Finance AI is shifting from copilots and governance concepts into production execution, where agents can retrieve data, draft entries, reconcile accounts, and even initiate payments under tightly scoped permissions. That matters because the bottleneck is no longer model capability but control architecture: identity, least-privilege access, approvals, logging, and exception handling. The next phase of finance automation will be defined by which teams can safely let agents act inside core systems.