.jpg)
We curate a lot of information daily. Our ongoing collections hold 100 article links, which amounts to about a week’s worth.
Every week, we select some of the articles to share in more depth.
TechTalks Substack - An AI system that is more 'human-like'
This research is particularly compelling because it directly confronts a persistent challenge: while current AI can perform superhuman feats, its operational logic is often incompatible with human cognition. They can be efficient but fail in unexpected ways, and their "black box" nature makes it difficult to understand their reasoning or predict their boundaries. Sakana AI's work on the CTM aims to bridge this gap.
The Algorithmic Bridge Substack - How Google Created an AI That Improves Itself
[AlphaEvolve is] an evolutionary coding agent powered by large language models for general-purpose algorithm discovery and optimization. AlphaEvolve pairs the creative problem-solving capabilities of our Gemini models with automated evaluators that verify answers, and uses an evolutionary framework to improve upon the most promising ideas.
InfoWorld - LiteLLM: An open-source gateway for unified LLM access
The growing number of large language models (LLMs) from various providers—Anthropic, Google, Meta, Microsoft, Nvidia, OpenAI, and many others—has given developers a rich set of choices but also has introduced complexity. Each provider has its own API nuances and response formats, making it a challenge to switch models or support multiple back fends in one application. LiteLLM is an open-source project that tackles this fragmentation head-on by providing a unified interface (and gateway) to call more than 100 LLM APIs using a single, consistent format.
Computerworld - Meta hits pause on ‘Llama 4 Behemoth’ AI model amid capability concerns
Behemoth was never intended to be just another model in Meta’s Llama family. It’s intended to be the crown jewel of the Llama 4 series, designed as a “teacher model” for training smaller, more nimble versions like Llama Scout and Maverick. Meta had previously touted it as “one of the smartest LLMs in the world.”
Hybrid Horizons: Exploring Human-AI Collaboration Substack - While We Debate Prompts, AI Is Building Itself
AlphaEvolve isn't just another Large Language Model spitting out text. It's an "evolutionary coding agent" that generates computer programs to solve complex problems, tests how well those programs work and then iteratively refines them to create better solutions. Think of it as a tireless team of programmers and mathematicians working 24/7, exploring countless possibilities that humans would never have time to try.
Forbes - After Reaching AGI Some Insist There Won’t Be Anything Left For Humans To Teach AI About
AGI is AI that is considered on par with human intellect and can seemingly match our intelligence. ASI is AI that has gone beyond human intellect and would be superior in many if not all feasible ways. The idea is that ASI would be able to run circles around humans by outthinking us at every turn.
Dark Reading - DeepSeek, Deep Research Mean Deep Changes for AI Security
Unlike traditional models that produce outputs directly from static weights, agentic models build intermediate reasoning steps that guide their final responses. While this enables more sophisticated and context-aware outputs, it also introduces new attack vectors. Adversaries can exploit these intermediate processes by launching chain-of-thought injection attacks. In such scenarios, a carefully crafted input or prompt might subtly alter the model's internal reasoning, nudging it toward a desired, potentially harmful outcome or exposing sensitive internal logic
AiThority - Unlocking AI Trust: How RAG and GraphRAG are Transforming GenAI
These efforts are particularly focused on areas such as content creation, content customization, customer self-service, knowledge discovery, knowledge management, intelligent search, and assisting customer service staff. This highlights the diverse range of applications where organizations are seeking to leverage the power of LLMs to improve efficiency and knowledge discovery and accessibility.
VentureBeat - MCP and the innovation paradox: Why open standards will save AI from itself
You’ve probably read a dozen articles explaining what MCP is. But what most miss is the boring — and powerful — part: MCP is a standard. Standards don’t just organize technology; they create growth flywheels. Adopt them early, and you ride the wave. Ignore them, and you fall behind. This article explains why MCP matters now, what challenges it introduces, and how it’s already reshaping the ecosystem.
Agora Substack - The Application Layer
In emerging AI verticals, we can expect a “winner (or buyer)-take-most” dynamic to unfold. The first application that truly nails the use case with a delightful UX can rapidly capture a huge share of users in that domain.
Forbes - Why AI Agents Will Trigger The Biggest Workplace Revolution In 25 Years
That’s an astronomical jump in market potential — and it's already manifesting in real business outcomes. Benioff pointed to 1-800-ACCOUNTANT, which just completed tax season with 70% of customer service inquiries handled autonomously without human interaction. Even Salesforce itself is seeing dramatic changes: "We have about 9,000 support agents, but they're doing a lot less work lately 'cause we have help.salesforce.com, which is this agentic layer that is resolving issues without human interaction."
VentureBeat - Alibaba’s ‘ZeroSearch’ lets AI learn to google itself — slashing training costs by 88 percent
The technique, called “ZeroSearch,” allows large language models (LLMs) to develop advanced search capabilities through a simulation approach rather than interacting with real search engines during the training process. This innovation could save companies significant API expenses while offering better control over how AI systems learn to retrieve information.
Computerworld - Perplexity AI’s quiet coup
This just-give-me-the-answer idea is perfect for a future in which people will ask an always-present assistant for the kind of information they used to get from Google. Smart glasses, of course, will be the main interface to arbitrary information, but other wearables, mobile devices, IoT devices and general purpose computers will enable a personal assistant that knows all about the world and knows all about us, individually and personally.
The Economy of Algorithms Substack - The Real Risk of AI Isn’t Replacement. It’s Irrelevance
Increasingly, jobs aren't being lost to more AI-savvy colleagues. They're disappearing because the system has simply stopped asking for them. Not out of malice or mismanagement, but because the way we create and capture value is shifting. Quietly. Systemically.
VentureBeat - Meet the new king of AI coding: Google’s Gemini 2.5 Pro I/O Edition dethrones Claude 3.7 Sonnet
The new version powers feature development in apps like Gemini 95, where the model helps match visual styles across components automatically. It also enables workflows like converting YouTube videos into full-featured learning applications and crafting highly styled components—such as responsive video players or animated dictation UIs—with little to no manual CSS editing.
Fast Company - ‘AI is already eating its own’: Prompt engineering is quickly going extinct
“AI is already eating its own,” says Malcolm Frank, CEO of TalentGenius. “Prompt engineering has become something that’s embedded in almost every role, and people know how to do it. Also, now AI can help you write the perfect prompts that you need. It’s turned from a job into a task very, very quickly.”
Network World - IBM wrangles AI agents to work across complex enterprise environments
The platform includes integration with more than 80 enterprise applications from providers such as Adobe, Microsoft, Oracle, Salesforce Agentforce, SAP, ServiceNow, and Workday. Pre-built agents with specific skills are available for certain business functions, beginning with HR, sales, and procurement. There are plans for additional domains, such as customer care and finance, in the coming months, according to Gunnar.
AI Adoption and Usage
Diginomica - CamundaCon – proof (again!) of why enterprises need to focus on AI strategic value over cost-cutting
Also, it's important to level set expectations with the new crop of tools. Mostly, enterprises are doing a lot of pilots and trying to orchestrate some processes. Forrester believes that a maximum of 1% of core business processes will be orchestrated by generative AI this year. A core issue is trust, since these systems tend to hallucinate, and it's hard to troubleshoot bias. Schaffrik says:
AI Changes Everything Substack - The Era of AI Adoption
92% of these early adopters report positive returns. The majority who quantified their ROI see an average 41% return — a figure that's leading them to increase investment across data infrastructure (81% of early adopters), LLMs (78%), supporting software (83%) and talent (76%).
Forbes - AI Is Not A Tech Upgrade—It’s A Business Transformation
Once this vision is in place, establish AI literacy across your entire organization. This should include not only hiring people who have a deep knowledge of future technology but also upskilling people across your organization and ensuring they are aware of the changes ahead. All employees should have at least basic knowledge, but there should be a few within each unit who possess a deeper understanding of what else can be achieved with the technology for the benefit of the business.
CIO Influence - Enterprise AI Adoption: Without the Hype
Enterprise AI won’t begin with bespoke, in-house models trained from scratch—it’ll piggyback on SaaS providers and ISVs using pre-trained, third-party large language models (LLMs). Companies like Salesforce, ServiceNow, Glean, or Palantir will embed generative AI (e.g., content creation) and agentic AI (e.g., task automation) into their platforms, delivering ready-to-use solutions. Picture a customer service chatbot that drafts responses, a marketing tool that generates campaigns, or a coding assistant that speeds up development—all powered by models from OpenAI, Anthropic, Gemini, or xAI, fine-tuned for enterprise needs.
ZDNet - AI agents bring big risks and rewards for daring early adopters, says Forrester
While many leading tech companies have recently introduced agents and businesses have begun implementing such systems into their workflows, Forrester predicts it will take a couple of years or more before the tech is safe and dependable enough for mainstream adoption.
ITPro Today - Putting AI Agents to Work for Your Business
According to Gartner's Top Strategic Technology Trends for 2025, by 2028, 33% of enterprise software will include agentic AI, up from less than 1% in 2024. These agents are expected to reach billions in operation and are projected to autonomously handle 15% of daily work decisions. Once theoretical, it's clear AI agents will play a crucial role in the future of work and in accelerating businesses' digital transformation.
Computerworld - Q&A: Ernst & Young exec details the good, bad and future of genAI deployments
Given the AI hallucinations, output errors, organizational data fragmentation and a lack of skilled IT talent to manage it, corporate leaders have good cause to be pensive. Yet, there’s not likely a consultant worth their salt that would advise an organization to not explore AI’s efficiencies, cost-savings and production-enhancing capabilities.
ZDNet - Tech leaders are rushing to deploy agentic AI, study shows
Most tech leaders surveyed (58%) also said their organizations were ahead of the competition on AI adoption. This result should be taken with a hearty grain of salt, however, as there's an obvious incentive to espouse such claims publicly, even in the absence of hard evidence: "this is more perception than reality," EY Americas Technology Sector Growth Leader Ken Englund said in a statement, "since these companies tend to have a higher opinion of their progress than is statistically possible."
SmartR AI Substack - How AI is Streamlining Compliance in Financial Services
According to Jas Dillion, Finance IQ, who led the implementation, "There are dozens of use cases inside each big financial services company right now that could benefit from automation with AI." Dillion emphasizes that "The pay back is often measured in weeks not years."
Forbes - 10 Key Findings From AWS Generative AI Adoption Index
While 90% of organizations now deploy generative AI tools, 44% have advanced beyond early testing to production deployment. Organizations conducted an average of 45 AI experiments in 2024, but only 20 will reach end-users by 2025, highlighting implementation challenges.
eWeek - AI Adoption Barrier That Managers Might Overlook, Says Duke Researchers
The Duke team conducted four experiments involving more than 4,400 participants, analyzing attitudes from both AI users and those who observed their behavior. The results showed a consistent bias: Individuals who received help from AI were judged more harshly than those who did not.
AiThority - Qlik AI Council: AI That Can’t Be Trusted Can’t Be Scaled—And AI That Can’t Be Scaled Is Just Theater
Despite record AI investment, most enterprises remain stuck in the lab. According to recent IDC research, while 80% plan to deploy agentic AI workflows, only 12% feel ready to support autonomous decision-making at scale. Trust in outputs is eroding amid growing concerns around hallucinations, bias, and regulatory scrutiny. And as models become commoditized, competitive advantage is shifting—not to those with the most advanced models, but to those who can operationalize AI with speed, integrity, and confidence.
SiliconANGLE - In the race to deploy AI, industry leaders confront hype and expense – but project a promising future
Some company executives and AI leaders are becoming concerned that the rush to build AI infrastructure and show results from that investment quickly is raising expectations beyond what the technology can reasonably deliver. Even Meta Platforms Inc. founder and Chief Executive Mark Zuckerberg (pictured) hedged slightly when asked whether AI could indeed be a bubble and might take longer to show significant results than people expect.
VentureBeat - How The Ottawa Hospital uses AI ambient voice capture to reduce physician burnout by 70%, achieve 97% patient satisfaction
Typically at the end of the day or shift, physicians have to then go back and finalize documentation from patient visits, he explained. But the tool has reduced after-hours, charting and documentation work for “all categories of physicians.” This not only saves them time but helps reduce burnout because they have less tedious work to do.
High ROI Data Science Substack - How To Accelerate AI’s Time To Value
Evaluate the vendor’s AI product and platform roadmaps, looking 1-2 years ahead to ensure they align with the business’s needs. Capabilities maturity is a feature of the business’s AI strategy. That means the business’s internal needs will advance over time, and the worst thing that can happen is to get locked into a solution that isn’t maturing in the same direction. I advise clients to look for solutions that run about a year ahead of the business.
VentureBeat - 5 strategies that separate AI leaders from the 92% still stuck in pilot mode
A new study from Accenture, out this week, provides a data-driven analysis of how leading companies are successfully implementing AI across their enterprises. The “Front-Runners’ Guide to Scaling AI” report is based on a survey of 2,000 C-suite and data science executives from nearly 2,000 global companies with revenues exceeding $1 billion. The findings reveal a significant gap between AI aspirations and execution.
Diginomica - Why governance musts be key to navigating the agentic AI imperatives - important learnings from new data from Gravitee
Operational efficiency, customer experience, and reduced costs were top drivers. Enterprises were also struggling with integration, privacy and security challenges. Controlling the costs of large language model (LLM) interactions stood out as the top concern. The most common approach was establishing a dedicated agentic AI team as a cross-disciplinary specialty, while nearly as many organizations turned to dedicated data science or engineering teams. About half of these efforts were backed by new budgets.
ZDNet - Why smart businesses use AI to offload tasks and supercharge their teams
Productivity gains will be immediate, but the overall potential for efficiency and effectiveness based on AI agent deployments requires additional analysis. Adoption of AI agents will reshape organizational structures. Over three-quarters (77%) of CHROs believe AI agents will transform organizational structures. Integrating AI agents is anticipated to create significant opportunities for human talent. Most CHROs (89%) believe AI and digital labor will enable them to transition employees into new, more impactful positions.
AiThority - How AI Has Redefined Supply Chain Efficiency and Traceability
AI is now being used to optimize everything from carbon footprint tracking to dynamic pricing strategies in retail. AI is also playing a key role in route optimization, helping retailers reduce unnecessary fuel consumption and waste. By analyzing real-time data on traffic, weather, and shipment priorities, AI is able to predict the most efficient delivery routes, reducing delays and cutting unnecessary fuel consumption. In warehouses, AI dynamically adjusts inventory levels to reduce overstocking and prevent waste, creating a more sustainable supply chain.
Computerworld - Will genAI businesses crash and burn?
This isn’t just cranky, cynical old me. In a recent IBM survey of 2,000 CEOs, Big Blue found that “only 25% of AI initiatives have delivered expected ROI [Return on Investment] over the last few years.” Further, “just over half (52%) of CEO respondents say their organization is realizing value from generative AI investments beyond cost reduction.”
ZDNet - Most CEOs find their C-suite lacks much-needed 'AI-savvy'
"AI is a step change in how business and society work. A significant implication is that, if savviness across the C-suite is not rapidly improved, competitiveness will suffer, and corporate survival will be at stake,"
ZDNet - Only 8% of Americans would pay extra for AI, according to ZDNET-Aberdeen research
In fact, the survey from March 2025 found that 71% of Americans would not pay extra for AI assistant features in the products that they use. And while this number does change based on age group (with the number rising to 81% for those 55+ and falling to 56% for those 18-34), there is little enthusiasm across the board for AI assistant features, with only 16% of the Gen Z demographic saying that they would pay extra for AI.
Computerworld - Real-world use cases for agentic AI
According to Gartner, agentic AI is the top strategic trend of 2025. By 2029, 80% of common customer services issues will be resolved autonomously, without human intervention. The firm also predicts that 33% of enterprise software applications will include agentic AI by 2028, and 15% of all day-to-day work decisions will be made autonomously.
We update our ongoing collections starting at 8:00 am MST.
Here are the direct links: