\n\tThis week we hear from Maine Pointe, the supply chain consultancy, which is breaking free from parent company SGS and is using AI to power its new, independent future. In other news, two companies, Cloudflare Inc. and Onton, are tackling the issue of trust and authenticity in the agentic commerce space, albeit in different ways.<\/p>\n
\t\tLeaving the nest\t<\/p>\n
\n\tMaine Point, the global supply chain and operations consulting firm, is now on its own. The company was acquired from SGS, and the deal was led by Joseph Esteves, Maine Pointe\u2019s chief executive officer.<\/p>\n
\n\tTerms of the deal were not disclosed.<\/p>\n
\n\t\u201cThe transaction establishes Maine Pointe as an independent, management-owned firm and gives the company greater flexibility to invest in its people, technology and client delivery model,\u201d the company said in a statement, adding that it will now \u201cmove faster and with greater agility, helping chief executives, private equity sponsors and operating leaders accelerate measurable improvements in profitability, cash generation, operational performance and enterprise value.\u201d<\/p>\n
\n\tThe company said it has over 20 years of implementation experience with AI capabilities along with a highly scalable delivery model. Maine Pointe said it has learned that AI creates the most value when it is combined \u201cwith deep industry and operational expertise and tied directly to a company\u2019s organizational and financial goals.\u201d<\/p>\n
\n\tEsteves said the company\u2019s relationship with SGS<\/a> \u201cwas an important and positive chapter in Maine Pointe\u2019s growth, and we are grateful for the support, global perspective and opportunities it provided,\u201d said Esteves.<\/p>\n \n\t\u201cAs an independent company, we can now move faster, invest more aggressively in AI and digital innovation, expand our world-class talent base and stay completely focused on one thing: helping our clients drive rapid, measurable results,\u201d the CEO added.<\/p>\n \n\tThe company said the buyout provides full continuity of its business and existing client teams and contractual commitments as well as performance guarantees.<\/p>\n \n\t\u201cWhat remains unchanged is our responsibility to work alongside clients and deliver measurable results,\u201d Esteves said. \u201cWhat changes is our ability to make decisions with greater speed, focus and ownership.\u201d<\/p>\n \t\tIt\u2019s all about trust\t<\/p>\n \n\tOnton\u00a0is launching Ontology 1, a trust and authenticity model for agentic commerce, which involves AI agents<\/a> to research, recommend and even make purchases for consumers. And therein lies the problem: when agents search, they could be finding data on products that are synthetic or manipulated and not authentic.<\/p>\n \n\tThe company said the agent has no reliable way \u201cto distinguish signal from noise and proceeds to recommend anyway.\u201d Shoppers end up buying based on information that was never trustworthy to begin with. This is where Ontology 1 helps.\u00a0<\/p>\n \n\tOnton said the model was built from scratch. \u201cIn\u00a0head-to-head benchmarks\u00a0against the most dominant players in product discovery, including Google Shopping<\/a> and Amazon, Onton\u2019s model outperformed on accuracy across every dimension tested,\u201d the company said in a statement. \u201cThe gap is sharpest in a domain where accuracy has historically been hardest to achieve: interpreting the veracity of product information.\u201d<\/p>\n \n\tAlex Gunnarson, co-founder of Onton, said the market is focused on building smarter agents. \u201cWe\u2019re focused on a different question: what should those agents trust?\u201d he said. \u201cAs commerce moves from clicks to conversations and eventually autonomous actions, the quality of information underneath those decisions becomes critical infrastructure.\u201d<\/p>\n \n\tOnton solves this problem by assessing both the product itself and the reliability of its surrounding data. By decoding the reasoning and visual cues behind user preferences, the company said the platform converts these inputs into valuable insights rather than discarding them as clutter. As a result, AI agents deliver shopping recommendations rooted in verified data.<\/p>\n \n\tAlongside its launch, Onton released research revealing that current top AI models struggle to filter out fake content and biased reviews\u2014proving that today\u2019s systems simply aren\u2019t built for modern, noisy web environments.<\/p>\n \n\t\u201cWe believe the next major internet platform will not be defined solely by who has the best model,\u201d said\u00a0Zach Hudson, co-founder of Onton. \u201cIt will be defined by who can provide the most trustworthy foundation for those models to operate on.\u201d<\/p>\n \t\tMaking safe purchases\t<\/p>\n \n\tCloudflare Inc. is also addressing issues of safety and trust in agentic commerce. The connectivity cloud company said Tuesday that it is rolling out Cloudflare Wallets and Cloudflare.pay, which will give AI agents that are deployed on Cloudflare \u201ca stable identity and the ability to make purchases online<\/a> safely within limits set by their human creators.\u201d<\/p>\n