Software
A GFT Technologies survey of 945 CIOs and CTOs reveals that legacy technology systems are blocking AI adoption, with more than 4 in 5 tech leaders canceling at least one AI pilot. 93% believe not modernizing legacy systems while using AI could eventually trigger an enterprisewide security event. Experts advise methodical modernization focused on business goals, as most enterprises are at varying stages of modernization.
Quantum computing is no longer a distant prospect, with the market reaching $1.4 billion in 2025 and expected to more than double by 2028. While most enterprises won't buy a quantum computer, CIOs should prepare by evaluating potential use cases and addressing cybersecurity implications. Experts recommend assigning a point person to explore applications and leveraging third-party vendors for access. The National Institute of Standards and Technology has set a 2035 deadline for federal contractors to migrate to post-quantum encryption standards, and Gartner predicts a 25% chance that a state actor will have a cryptographically relevant quantum computer by 2029, rising to 75% by 2031. CIOs must inventory encrypted assets and engage vendors on quantum readiness.
A report released by McKinsey this week shows that large enterprises are expanding agentic AI applications faster than small and medium-sized enterprises, with usage growing rapidly. Based on a survey of over 1,700 employees, the report found that nearly a third of respondents gave up purchasing at least one software feature because they could build software themselves. Although agent tools have higher computing costs, technology leaders still see value and plan to increase investment.
Walmart's Chief People Officer, Donna Morris, revealed at the MIT Technology Review EmTech AI Summit that the company plans to provide AI skills training to 2.1 million employees over the next few years, covering all roles from technical teams to store greeters. Currently, AI certifications for 1.7 million employees in the U.S. and Canada are now available, with training aimed at integrating agentic AI into employee workflows to reduce friction and enhance customer experience. Morris emphasized that AI adoption has not led to job reductions, with the company's workforce size remaining the same as six years ago, while revenue has grown significantly.
The software industry is undergoing disruptive changes driven by AI. Research released by Forrester this month shows that AI will replace or weaken existing software platform features. CIOs need to look beyond short-term financial impacts, review their vendor portfolios from a long-term perspective, identify commoditized features, and address the risk of uncontrolled AI spending.
In its Q2 earnings call, SAP reaffirmed the central role of its AI strategy, with cloud revenue up 24% year-over-year and total revenue reaching $11.24 billion. CEO Christian Klein highlighted ERP migration as a growth driver and emphasized the vision of the autonomous enterprise. The company launched the SAP Business AI Platform and Autonomous Suite, acquired Dremio and Reltio to strengthen its data foundation, and expanded collaborations with Google Cloud, Microsoft, Snowflake, and Nvidia. Internally, it is also advancing AI talent recruitment and employee skills transformation, aiming to cover over 90% of its workforce.
In the quarter ending June 30, IBM's infrastructure segment revenue fell 7% year-over-year, with IBM Z mainframe business revenue down 42% year-over-year. The company attributes this to customers postponing capital expenditures due to chip shortages and rising hardware prices, rather than customer attrition. CEO Arvind Krishna stated that one-third of deferred deals were completed within three weeks after the quarter ended, indicating demand remains.
Gartner's latest forecast shows that global end-user spending on AI models and platforms will jump 63% from last year's base, reaching $64 billion. As costs rise, CIOs are more strictly reviewing AI budgets, focusing on usage efficiency, cost control, and measurable results. Analysts point out that agentic AI drives most model spending to be embedded in platforms, and model vendors are transforming into platform intelligence providers, with competition shifting toward enterprise end-to-end workflows.
Against the backdrop of agentic applications accelerating AI usage growth, the difficulty for enterprises to control AI costs has not been alleviated by falling token prices. Flexera's survey shows that about three-fifths of IT professionals report an increase in AI overspending, and more than two-thirds lack visibility into AI software usage. In a blog post on Tuesday, OpenAI proposed five recommendations to help enterprises manage spending and understand AI usage and value creation in the agentic era.