On October 20, 2026, Salesforce said it had signed a definitive agreement to buy Braze, a customer engagement platform, for approximately $5.66 billion in enterprise value, paying $47.50 in cash for each Braze Class A share.1 Five days earlier, ServiceNow unveiled a product meant to answer a different question: once companies let software agents act on their own, what stops those agents from doing things they should not?3 Together the two announcements show how business software is being reorganised around autonomous AI agents. The market is consolidating, and vendors are racing to sell control as much as capability.
The deal in proportion
$5.66 billion is hard to picture, so here is Salesforce's own scale. In its fiscal 2025 results, drawn from its SEC filing, Salesforce reported revenue of $41.525 billion, up from $34.857 billion a year earlier.2 By our arithmetic, the Braze price is about one-seventh of one year's revenue. It is also roughly 38% of the $14.996 billion in cash the business generated from operations in fiscal 2025.2
One caution on the comparison. 'Enterprise value' is a headline measure of what a company is worth. It is not the same as a cheque written from the buyer's bank balance. The announcement does confirm that Braze shareholders are being paid in cash.1 Salesforce reported cash of $7.327 billion at the end of fiscal 2025, down from $8.472 billion a year earlier.2 A deal of this size will therefore not be funded from idle money alone. The dossier does not say how the purchase will be financed, and we do not speculate.
Braze is meant to bring Salesforce 'real-time engagement', and the company describes it as a platform that responds to customer behaviour across mobile apps, websites, email, SMS and WhatsApp.1 Salesforce, which calls itself 'the world's #1 AI CRM', is the same company behind Agentforce, Data Cloud and Einstein, according to our entity records.9 Buying a system that reacts to customers in real time fits an agent-led strategy. The announcement itself, however, is where that framing comes from, so treat it as the company's account.1
What Salesforce's checked numbers show
The financial backbone here is verified. These figures are canonical, taken from Salesforce's SEC filings, and fidelity-checked.
Fiscal 2025 revenue rose from $34.857 billion to $41.525 billion.2 Net income climbed faster, from $4.136 billion to $7.457 billion, and operating income from $5.011 billion to $8.331 billion.2 Gross profit rose from $26.316 billion to $32.255 billion, while the cost of delivering its products went from $8.541 billion to $9.270 billion.2 By our arithmetic, that lifts gross margin from about 75% to about 78%. In plain terms, Salesforce grew and became more profitable at once.
Spending priorities are visible too. Research and development spending increased from $4.906 billion to $5.993 billion, while capital expenditure, the money spent on buildings and equipment, fell from $736 million to $594 million.2 The company is investing in software development more than in physical assets. That is typical of a software business, and the numbers confirm it. Operating cash flow rose from $10.234 billion to $14.996 billion.2 If you hold a broad stock-market index fund, you may well own part of Salesforce, but the dossier gives no index weighting and we will not guess one.
The new product: a passport for software agents
ServiceNow's announcement is the clearest statement of the governance problem. It launched AI Passport, described as an authority layer and a cryptographically signed credential. It verifies the scope of both in-house and third-party AI agents in real time and gives runtime control over what agents and MCP servers are authorised to do.3 It also widened its Super Intelligence Control Tower, adding discovery, visibility, control and cost-to-value measurement on a single data model.3
ServiceNow's Jon Sigler framed the logic: 'As AI agents proliferate across the enterprise, visibility alone isn't enough. Enterprises need confidence that every AI agent is operating within its authorized scope.'3 That is a vendor describing its own product, so it is a sales pitch as much as an analysis.
An independent-sounding argument for the same idea comes from a sponsored piece published by VentureBeat and presented by EDB. EDB wrote: 'The enterprise should not rely on a model choosing to follow policy. The policy has to be enforced by the system. That is the difference between hoping an actor stays in bounds and constructing bounds it cannot cross to begin with.'4 Because the article was presented by EDB, which sells database technology, it is also the view of an interested party.
Via News's analysis of the data suggests the shared message is the story. Vendors with very different products are converging on one claim: for regulated sectors such as finance, controls that are built in and enforced matter more than the cleverness of the agent. The dossier's topic narrative also lists SAP's 'Autonomous Enterprise', Zeta's AthenaOS and its own inference model, and Palantir partnerships in banking, aviation and sovereign AI infrastructure.10 It holds no primary documents on those, so we do not assess them here.
An investor's warning about maths
The cautionary voice comes from Sean Jacobsohn of Norwest, speaking to Crunchbase News. On finance software he was blunt: 'You don't want AI doing calculations because it is not good at math. There are certain workflows it can handle where it doesn't produce precise numbers. But when you need precision, accuracy and calculations, you can't rely on AI for that.'5 For anyone whose pension or savings depends on a company's books being right, that is worth keeping in mind next to every promise of 'autonomous' finance.
Jacobsohn also sees where disruption is plausible. 'I think it would be very hard to disrupt the core products of Workday, ADP, SAP, UKG and Dayforce. But it's easier to disrupt some of their secondary products, where the category isn't their core business.'5 He adds that when a chief financial officer is the direct buyer, 'There's actually one less layer of approval when they're buying their own software, so it is a little easier to replace it when they're the direct buyer.'5 His test for founders is practical: 'Before I invest, I'll go on a lot of sales calls I set up with the CEO to see how good they are at selling.'5 Crunchbase News is rated at 50% on our source-reliability measure, with 50% of 38 checked claims holding up. The quotes are his own words, but they are opinions, not data.
The small end of consolidation
Consolidation is happening below the giants as well. On October 6, 2026, Tiny Ltd. completed its acquisition of Oso Cloud, an authorization and access-control software business, for approximately $1.9 million in cash.6 Tiny says Oso adds about $5.3 million in annualised recurring revenue, lifting its total by about 7.6%, from $70.0 million to about $75.4 million on a pro forma basis.6 Oso served approximately 80 customers in security, fintech and developer software, all on recurring subscription contracts, several of them multi-year.6 The price is roughly a third of a year of that recurring revenue. The point is relevant to this story because Oso's specialism is authorization, the very control function ServiceNow and EDB are describing, and a small holding company has now bought it.
Not every deal runs smoothly. On October 20, 2026, ZenaTech sent a formal complaint to the BCSC and the TSXV asking for regulatory intervention over Boardwalktech's failure to hold its 2026 annual general meeting and what it calls continuing governance deficiencies. In an open letter to shareholders it also criticised board independence and repeated small dilutive financings.11 These are ZenaTech's allegations. The dossier contains no response from Boardwalktech, so they remain unproven.
Agents in the field
Real deployments are modest but growing. Dextr AI says its hotel agents now handle more than one million interactions a month across hundreds of hotels.7 In procurement, CEO Vikas Gupta says Dentira provides procure-to-pay services for multi-site healthcare groups, with clients supporting hundreds, sometimes thousands, of clinics. He adds: 'Dentira is already the biggest player in a few verticals, and we're planning to become the largest vertical software provider in procurement technology overall.'8 Both are company self-descriptions. For the Dextr piece, measured reliability of its publisher is 50% of 38 checked claims, so the usage figure should be read as the company's claim, not an audited number.
How reliable is the news behind this?
The two big announcements reached us through NewsEOD, which carries company press releases. On our measurement, 56% of 5,087 checked claims from that source held up.1 That is why we lean on SEC-filed numbers for Salesforce's finances and treat deal terms as the companies' own statements until completion filings confirm them.
What to watch
- Whether the Braze deal closes on the stated terms of $47.50 a share, and how Salesforce funds it, since its fiscal 2025 cash balance was lower than the headline price.12
- Whether governance tools such as AI Passport are adopted by regulated customers, or remain a marketing claim.3
- Whether ZenaTech's complaint draws a regulatory response or a reply from Boardwalktech.11
- Whether Jacobsohn's warning about AI and calculation shapes how finance software is actually sold.5
Sources:
1 NewsEOD, "Salesforce Signs Definitive Agreement to Acquire Braze" (2026-10-20)
2 Via News verified data — Salesforce revenue, net income, operating income, gross profit, cost of revenue, R&D, capex, operating cash flow, cash; SEC filing (FY 2024 and FY 2025)
3 NewsEOD, "ServiceNow advances its operational harness for enterprise AI with AI Passport and expanded governance capabilities" (2026-10-15)
4 VentureBeat AI, "When agents act on their own, governance has to live in the data layer" (undated)
5 Crunchbase News, "The Sales Test This Norwest Partner Gives Founders Before He’ll Invest" (undated)
6 NewsEOD, "Tiny Completes Acquisition of Oso Cloud" (2026-10-06)
7 Crunchbase News, "Exclusive: From Booking Calls To Late Check-Ins, Dextr AI Raises $6.7M For Hotel AI Agents" (undated)
8 CB Insights, "CEO Interview: Dentira" (undated)
9 Via News entity graph — Salesforce: Agentforce, Data Cloud, Einstein (developed by Salesforce)
10 Via News topic narrative — "Enterprise Software Races to Govern and Consolidate Around Agentic AI" (no primary source document in dossier)
11 Via News event record — "ZenaTech requests BCSC and TSXV intervention" and "ZenaTech open letter to Boardwalktech shareholders" (October 20, 2026)
