tag:dlvr.it,2018-03-23:ac6861e9e2d249fc92b2895b5ff7a2e0 2026-08-30T21:47:25-07:00 tag:dlvr.it,2026-08-30:3a78926971db55307cdff6d16d83e172 2026-08-30T21:47:25-07:00
"Full stack platforms are replacing processor-driven themes. Companies are looking to build quantum ecosystems. Commercialization is progressing but it is uneven. Quantum computing revenue depends on government customers and experimentation in key industries with defined use cases. Sovereign quantum computing is seen as critical and vendors are looking at supply chain control and manufacturing as strategic assets. Applications are focused on optimization and scientific workloads. Hybrid deployments combining quantum computing, AI and supercomputing are emerging. " ]]> tag:dlvr.it,2026-08-30:59f71e2fe576d73ef8899f229583c07c 2026-08-30T21:47:25-07:00
"Across all six risk types, the enterprise must constantly ask whether each AI project is developing a durable asset or renting a capability that will be commoditized within a short period. Both answers can justify funding. However, they justify different amounts, on different terms, with different expectations of permanence. The distinction between investing to acquire capability and renting capability belongs on the business case record." ]]> tag:dlvr.it,2026-08-30:4644bace65b47ebc7fb731ed5f56b508 2026-08-30T21:47:25-07:00
"Palantir is increasingly positioning itself in this space through its Foundry platform, which layers an AI-enabled ontology over existing ERP investments rather than replacing them. Its partnership with Accenture and Deloitte frames this explicitly as a way for organizations to “reimagine work and infuse AI while rationalizing existing technology”, without waiting for a full ERP go-live. It’s a legitimately different value proposition, and one that more ERP program sponsors are asking about. Data remediation is becoming the longest pole in the tent on ERP transformations, and it rarely gets the program budget it deserves at the outset. " ]]> tag:dlvr.it,2026-08-30:638177dae974e5ba4d5c9ca437b6607f 2026-08-30T21:47:24-07:00
"In discussions with supply chain leaders, most believe that the global stock market is overheated with AI spending. Just as we have forgotten the lessons of the pandemic, few leaders remember the political drama of large stock market drops on the supply chain. Supply chain architectures aren’t built to plan easily when there are knowns and unknowns. The traditional focus is optimization of known inputs to known outputs. One of the best sources of demand data comes from transportation reporting. Here are some current trends. Ships sitting in the Strait of Hormuz tie up global capacity. The rise of global shipping volumes amid geopolitical friction—especially in Asia—has created trade imbalances. Ocean shipping reliability now hovers around 60-65%, down from pre-pandemic norms of 70-80%. Roughly 8% of global ocean container ship capacity is now unavailable due to shipping inefficiencies. As inefficiencies rise, costs escalate and lead times are more variable. Ocean container equipment could easily become a constraint. " ]]> tag:dlvr.it,2026-08-30:5a156be58aaf253186c4d3493f67b5ff 2026-08-30T17:42:44-07:00
"Jon opened with numbers rather than opinion, a habit more of us should copy. McKinsey's recent work on AI measurement found that nearly eight in ten companies are using generative AI in some capacity, while around sixty percent report not seeing enterprise-wide EBIT impact from those programs. The gap between activity and impact is apparently not closing. Instead, it seems to be widening. A small group of over-performers is automating end-to-end workflows inside specific domains and getting results, and even they argue about what to measure and how to attribute the improvement." ]]> tag:dlvr.it,2026-08-30:de9d0c6758068abb666a6870ae1bcd52 2026-08-30T17:42:44-07:00
"Fu and colleagues gave the effect a name in Not All Tokens Are Equal: token inflation, the gap between advertised per-token pricing and what a workflow actually consumes once it retries what it got wrong. They measure inflation as high as 4.25x on multi-hop question answering, and they show that FrugalGPT, one of the standard cost-aware routers, underestimates true expense by more than 2x on hard tasks. This mechanism is what both drives the cost and nobody prices in: a failed reasoning chain is not only a wasted call, it is a call that gets re-sent in full, with history attached, to a more expensive model." ]]> tag:dlvr.it,2026-08-30:d4b7fde7a10d369b761eefd8e2d903ff 2026-08-30T17:42:44-07:00
"Prompt injection is OWASP's No. 1 LLM risk, but ranks No. 12 in 6,639 real incidents — an independent analysis finds no statistically detectable agreement." ]]> tag:dlvr.it,2026-08-30:6e1d67db995a2b82ee70b5b46531e614 2026-08-30T17:42:44-07:00
"I recently encountered the same problem in a different form while reviewing an implementation SOW. The client had specifically requested contractual “off-ramps” because it wanted the ability to exit the engagement at defined points if the program was not working. When I asked an LLM to perform a general review of the agreement, it highlighted the off-ramps as a significant advantage for the client and characterized them as providing meaningful flexibility. On the surface, that conclusion made sense. Off-ramps sound client-friendly. The problem appeared when we examined how the provisions actually worked. The conditions surrounding the off-ramps made them extremely difficult to exercise in practice. More importantly, the provisions needed to be evaluated against the broader termination rights a client would ordinarily expect in a major engagement, including termination-for-convenience rights that had effectively been constrained by the structure." ]]> tag:dlvr.it,2026-08-30:a8ae49077cf0cf169bab411ff2ce4227 2026-08-30T16:35:08-07:00
"Yep, the silly season is upon us.... time to polish thos BS detectors! Get ready to decode over-the-top AI keynote messaging. Brian and Jon have your survival guide ready! Each will reveal the questions customers should be pressing, and how to get the most insights with the least event friction! We'll also be picking our top underrated stories of the month. As usual, Brian will have his infamous slide deck - bring your savviest and snarkiest commentary. We may even bring a surprise guest on cam, to help us all get ready for what pitfalls lie ahead..." ]]> tag:dlvr.it,2026-08-30:737ea559a745357c1f9277994572fd93 2026-08-30T16:35:07-07:00
"This speaks to a more general limitation of current AI systems. They are creative in the sense that they can recombine existing ideas in novel ways. But they are not creative in others: they have not yet developed conceptually new theories or structures. And while they have larger working memories than humans do, know more about more different things than any particular human does, and can process information faster than humans, can, true novelty is still largely beyond their reach."]]> tag:dlvr.it,2026-08-30:314f4e7d8ec0c916c4b42c8f6e735f2d 2026-08-30T16:35:07-07:00
Enterprise hits and misses - agents need meaning, not just data. Anthropic grapples with model pricing, and retailers get an economic gut check. https://ift.tt/b9E2wM1 This week - agents need meaning - and proper, contained actions. But how do we get there? Anthropic gets a taste of frontier model pricing backlash, and retailers run into macro-economic crosswinds. Oh, and the best enterprise phrase of the year is coined. Your whiffs are here, and so am I. via Jon Reed – diginomica https://diginomica.com August 24, 2026 at 03:16AM ]]> tag:dlvr.it,2026-08-23:80c9c19321066d138e59d76d80ff3dca 2026-08-23T16:57:03-07:00
"In the first post in this series, Modern Software Registries are a Trust Service, I traced how npm, PyPI, RubyGems, crates.io, and NuGet all converged on the same response to a year of supply chain attacks, issuing short-lived credentials tied to verified build pipelines rather than trusting long-lived tokens sitting in CI. My conclusion there was that registries are becoming policy engines as much as distribution systems. This post is about what happens when somebody works out how to charge for that. My argument begins by looking at hardened container images, where the trust-service model is furthest along. I then trace the same model in extension registries, where the Eclipse Foundation now sells an SLA on Open VSX and Microsoft’s Visual Studio Marketplace restricts extension use to Microsoft products. I argue the pattern across these markets is the same: the artifact is free, assurance is the product." ]]> tag:dlvr.it,2026-08-23:c498af93689c59c328cdf90a663e18bc 2026-08-23T16:57:02-07:00
"Enterprises won't sell their data because it's their competitive advantage--or so they think. The common thinking is that every enterprise will leverage its own data to train open models for agentic AI. Now data is the business moat, but it's highly likely that there will be enterprises that don't have the talent, skills or business acumen to leverage their own datasets. For now, companies are trying to follow the Data Inc. approach, which was outlined in 2023 by Constellation Research CEO R "Ray" Wang. " ]]> tag:dlvr.it,2026-08-23:7e65a7e52ef520d90c516b2f00af7233 2026-08-23T15:49:33-07:00
"Enterprise AI is in the pressure cooker: questionable results on the one hand, rising token costs on the other. But enterprise vendors like Oracle are pressing hard down a different path to value - one that reduces model costs and breaks away from frontier model hype. Can Oracle build a "system of outcomes"? And, speaking of building, can customers and partners build on Oracle's SaaS and AI foundations? In our latest Executive Intelligence podcast, Jon Reed airs out his concerns on enterprise AI, and Leone makes his case." ]]> tag:dlvr.it,2026-08-23:3265cb61594c410cb5f62c9437f3da49 2026-08-23T15:49:32-07:00
"AI researcher Simon Willison spent two days with Anthropic’s Fable AI, and called it “relentlessly proactive.” For example, he asked it to track down a stray scroll bar in a web app. He came back to find it had opened browsers, written its own screenshot tooling, created its own page to re-create the bug, and stood up a local web server to collect measurements. It found the bug and, along the way, did many surprising things he never asked it to do. And we are seeing similar behavior with all recent AI models when combined with flexible harnesses. This kind of behavior could easily go off the rails. Tell an AI agent to book you a flight and, finding the airline’s site says sold out, it might break into the booking database and force a reservation. Ask it to schedule a meeting and it might snoop your password to access your calendar. Tell it to save money on your phone plan and it might cancel the plan outright, or scam someone else into paying the bill." ]]> tag:dlvr.it,2026-08-23:35b22c7c5372f860c7221acdb91fdcb1 2026-08-23T15:49:32-07:00
"The concern that AI could weaken the next generation of professional expertise is real, but the outcome is not predetermined. If firms simply automate the work and allow the developmental experiences to disappear, the experience base may decline. If they treat AI as a way to manufacture realistic experience, compress learning cycles, and test judgment before people are placed into roles where clients bear the consequences, the opposite may happen. We may end up developing strong professionals faster than we ever have before." ]]> tag:dlvr.it,2026-08-23:392b9b7c2e43ca596ae2bd6194ad3393 2026-08-23T13:48:18-07:00
"Four of five enterprises that secured AI agent identities never built isolation to contain a compromised agent, VentureBeat's July Pulse survey found." ]]> tag:dlvr.it,2026-08-23:0e6113d163a9a98d78a11e6123eb6c77 2026-08-23T13:48:18-07:00
"Now we enter the era of Artificial Intelligence – again. The hype is deafening. We are told that Generative AI and LLM technology will revolutionize customer service. We are told that chatbots (err, agents) connected to a RAG architecture will flawlessly guide customers through their issues. Let me be absolutely clear. If your underlying data architecture is a mess, an LLM will simply hallucinate solutions based on that mess. Using RAG to query a broken knowledge base will just give you highly confident, grammatically correct wrong answers. Best regards from Air Canada! AI is not magic. It is a tool that accelerates whatever processes you have in place. If your process is designed to ignore five-sixths of the customer's reality, AI will just ignore them faster. True innovation in CX requires a solid architectural foundation. You need a unified data layer that provides a single, accurate view of the customer. You need integration across your entire technology stack. Your CRM must talk to your billing system. The billing system must talk to your support platform. Without this integration, your AI initiatives are doomed to fail." ]]> tag:dlvr.it,2026-08-23:831f48bdfc87a4c74a1a55766def9e81 2026-08-23T13:48:17-07:00
"For the first time, a majority of adults under 30 (55%) now say they’re more concerned than excited about AI. About one-in-ten say they’re more excited than concerned, and roughly a third feel both equally. Their concern is now on par with those in their 30s and 40s, and those 65 and up." ]]> tag:dlvr.it,2026-08-23:25439c64abcec0dec46ad87e6905356c 2026-08-23T12:40:48-07:00
"AI-generated code often looks immaculate. It passes linting, features clean syntax, and arrives with unit tests attached. But slop debt is dangerous precisely because it is invisible—it satisfies basic automated checks while embedding subtle semantic errors or invariant violations that only emerge under operational pressure. Even when the output is well-structured, generation scales dramatically faster than human comprehension. An agent can produce thousands of lines of implementation in minutes, but an organization still requires human judgment, domain context, and operational experience to verify that those lines correctly represent business invariants." ]]>