Good morning. Verizon sees AI not just as a tool to boost efficiency but as a potential new revenue engine.
I recently spoke with Verizon CFO Tony Skiadas, who discussed how the company is working to repurpose parts of its wireline network to support AI workloads and what that could mean for Verizon’s future.
Reimagining wireline for AI
Skiadas explained that Verizon (No. 30 on the Fortune 500) is testing ways to leverage its existing central offices—facilities largely freed up as copper networks are replaced by fiber—for AI processing power at the edge.
“Fiber takes up a fraction of the space compared with the old copper network,” Skiadas told me. That frees up facilities equipped with space, power, and cooling—exactly what’s needed to handle AI workloads at scale, he said.
The initiative, internally called Verizon AI Connect, centers on repurposing those assets rather than selling them. According to Skiadas, the combination of facilities and fiber positions Verizon to serve hyperscalers—large tech companies requiring custom AI infrastructure—with unique value. The company might have to spend some capital on its facilities, but it already has many of the assets in place to deliver these workloads at scale, he said.
“This is probably a medium- to long-term exercise,” he noted, “because every deal is highly customized.” Skiadas added: “But I like what I’m seeing from a sales funnel perspective. We talked about a billion-dollar sales funnel at the beginning of the year, and that’s actually doubled in terms of potential opportunity.”
While some smaller agreements could materialize this year, larger deals will take more time due to the complexity of building fiber or upgrading facilities, Skiadas explained. “It’s not a flip-the-switch thing,” he said. But the current level of demand is encouraging and will help guide where the company invests, he added.
AI inside Verizon
Beyond customer offerings, Verizon also is using AI internally to improve efficiency and service, Skiadas said. He pointed to AI-driven personalization in its customer plans, tools that help support agents find answers faster, and network optimization powered by machine learning.
AI is making Verizon’s customer care both more efficient and more effective, he said. “The customer is not waiting for 10 or 15 minutes for an answer.” Verizon is also applying AI in its network and across back-office functions to improve forecasting, accuracy, and decision-making, he added.
“I’m pushing my own team on this, too, to continue to innovate,” Skiadas said. “I even use it myself for simple things.” For example, he uses it to digest reports and summarize documents. “It’s a time saver for me,” he said. “And I tell people, if I can use it, anybody can. So that’s my motivation to my team.”
Regarding ROI of AI: “I think it’s going to take time,” Skiadas said. Some benefits, like productivity gains in customer care, are easy to quantify, while others—such as efficiency improvements in finance or better decision making—are harder to measure directly. The true measure, he emphasized, is how effectively Verizon employees can make forward-looking decisions. Ultimately, Skiadas sees the value of AI less in looking backward and more in improving forecast accuracy, guiding decisions, and enabling employees to focus on higher-value work.
I asked Skiadas what he thinks makes Verizon stand out among its competitors. Over the past seven years, Verizon has invested about $200 billion in wireless spectrum and networks—spending roughly $17–18 billion annually—to continually strengthen its network, Skiadas said.
“That’s really the hallmark of our company, and then giving customers choice and flexibility,” he said.
Raja Dakkuri, EVP and CFO of Cohen & Steers, Inc. (NYSE: CNS), has decided to resign from the company effective Oct. 17 after accepting another opportunity. Cohen & Steers has appointed Michael Donohue, SVP and controller, as interim CFO. The company has begun a search, considering both internal and external candidates, to find a permanent successor.
Hashim Ahmed has been appointed CFO of New Found Gold Corp. (NYSE-A: NFGC), effective immediately. Current CFO Michael Kanevsky will assist with the transition. Ahmed brings 25 years of experience, most recently serving as EVP and CFO at Mandalay Resources Corp. prior to its acquisition by Alkane Resources Ltd. He has also held CFO roles at Nova Royalty Corp. and Jaguar Mining Inc., and spent seven years at Barrick Gold Corp.
Big Deal
U.S. corporate bankruptcies climbed for a fourth straight month in August, according to S&P Global Market Intelligence data. Filings by large public and private companies rose to 76 from 71 in July. Year-to-date, 524 companies have filed through August, the most for the period since 2010. The data includes companies with public debt and assets or liabilities of at least $2 million or private companies with assets or liabilities of at least $10 million at the time of filing.
U.S. corporations reduced debt in the second quarter, according to S&P Global Market Intelligence, and could see further relief in the months ahead as the Federal Reserve is expected to resume cutting interest rates. “However, the impact from these cuts may be limited if yields for mid-dated and long-dated Treasurys do not decline alongside the Fed’s easing monetary policy,” the report states.
Courtesy of S&P Global Market Intelligence
Going deeper
“Trump wants to end a half-century-old mandate on how companies report earnings” is a Fortune report by Nino Paoli.
From the report: “In a Truth Social post on Monday, President Trump said companies should instead only be required to post earnings every six months, pending the U.S. Securities and Exchange Commission’s approval. This change would break a quarterly reporting mandate that’s been in place since 1970. ‘This will save money, and allow managers to focus on properly running their companies,’ Trump wrote. He added that China has a ’50 to 100 year view on management of a company,’ as opposed to U.S. companies required to report four times in a fiscal year. China’s Hong Kong Stock Exchange allows companies to submit voluntary quarterly financial disclosures, but only requires them to report their financial results twice a year.”
Overheard
“A well-designed digital identity system doesn’t just verify that you are who you say you are. It also protects your ability to limit what you reveal.”
—Will Wilkinson, director of government affairs for identity provider Persona, writes in a Fortune opinion piece titled, “America needs a digital identity strategy.”
This is the web version of CFO Daily, a newsletter on the trends and individuals shaping corporate finance. Sign up for free.
In a landmark move that signals a definitive shift in how major media conglomerates approach artificial intelligence (AI), OpenAI has gone from the company that had unapproved Disney princesses being made from its tools to a $1 billion partnership with the house of mouse itself. Disney CEO Bob Iger unpacked the deal jointly with OpenAI CEO Sam Altman in a TV interview with CNBC’s Squawk on the Street, explaining “we’d rather participate in the rather dramatic growth, rather than just watching it happen and essentially being disrupted by it.” He also reframed the issue of how AI is reshaping entertainment, business, even work itself: “Someone once said to me that creativity is the new productivity, and I think you’re starting to see that more and more.”
The deal, which brings Disney’s intellectual property to OpenAI’s video generation platform Sora, is structured to balance “aggressive” intellectual property protection with a willingness to embrace inevitable technological disruption, Iger said. Under the terms of the three-year agreement, Disney will license approximately 200 characters for use within Sora, allowing users to create short-form videos featuring iconic figures ranging from Mickey Mouse to Star Wars personalities.
Iger framed the partnership not as a concession to AI, but as a necessary evolution—and one that is actually good for human artists. This is because the deal does not include name and likeness, nor does it include character voices. “And so, in reality, this does not in any way represent a threat to the creators at all, in fact, the opposite. I think it honors them and respects them, in part because there’s a license fee associated with it.” Iger stressed repeatedly Disney wants to be on the cutting edge of how technology reinvents entertainment. “No human generation has ever stood in the way of technological advance, and we don’t intend to try.”
The partnership stands in stark contrast to Disney’s relationship with other tech giants. On the same day the OpenAI deal was announced, Disney sent a cease-and-desist letter to Google regarding alleged misuse of IP. Iger explained the divergence in approach by noting that, unlike Google, OpenAI has agreed to “honor and value and respect” Disney’s content through a licensing fee and safety guardrails. “We have been aggressive at protecting our IP, and we have gone after other companies that have not honored our IP,” Iger said, adding conversations with Google had failed to “bear fruit.”
A win-win partnership?
For OpenAI, reportedly under pressure from the aforementioned Google—whose Gemini 3 has been hailed by AI luminaries such as Salesforce billionaire Marc Benioff—the deal represents a validation of its generative video technology. Altman told CNBC user demand for Disney characters was “sort-of off the charts,” and he envisioned a future in which fans can generate custom content, such as a “Buzz Lightyear custom birthday video” or a personalized lightsaber scene. Altman argued the partnership would unlock “latent creativity” in the general public by lowering the skill and effort required to bring ideas to life.
The collaboration will also extend to Disney’s own streaming platform. Iger revealed plans to integrate “user prompted Sora-generated content” directly into Disney+. He said specifically Disney has “wanted for a long time to have what we will call user-generated content on our platform,” suggesting this partnership is a defensive move with regard to streaming giant YouTube and social media epicenter TikTok, which is partially under the control of the Ellison family that also controls entertainment rival Paramount.
The deal includes undisclosed warrants, giving Disney a financial stake in OpenAI’s success. Iger confirmed the warrants and declined to offer more specifics. He compared this forward-thinking approach to Disney’s 2005 decision to license shows to iTunes, viewing the OpenAI partnership as the modern equivalent of boarding a “profound wave” of societal change.
Iger revealed the groundwork for this deal was laid several years ago, saying he had first met Altman in 2022, when he was retired from Disney, before his comeback as CEO. Altman gave Iger a “bit of a road map” about where OpenAI was headed, and Disney has been “extremely impressed” with OpenAI’s growth since then, with all of Altman’s predictions from 2022 coming true a lot faster than either party realized. Iger added Disney sees great opportunities to license other product from OpenAI in the years ahead, which he sees being a huge push in “essentially accomplish[ing] a lot of what we feel we need to accomplish in the years ahead.”
For the third meeting in a row, the Federal Reserve cut interest rates—a “hawkish” move in an effort to help a softening labor market. The 0.25% cut brought the interest rate range to 3.5% to 3.75%—but economists and housing experts warn that’s not going to affect mortgage rates in the way potential homebuyers were hoping for.
Chen Zhao, head of economics research at Redfin, wrote in a Wednesday post that the Fed’s December interest rate cut won’t move mortgage rates “because markets have already priced it in.”
The Federal Reserve controls the Federal funds rate, which is a rate that banks charge each other and is more closely tied to credit cards, personal loans, and home-equity lines. A standard 30-year mortgage, on the other hand, is a long-term loan, and the pricing of those loans are tied more closely to yields on longer-term bonds like the 10-year Treasury and mortgage-backed securities.
“Since this rate cut was no surprise, the markets have taken it in stride,” 43-year mortgage industry veteran Melissa Cohn, regional vice president of William Raveis Mortgage, told Fortune. She said more dropping shoes in terms of economic data will be the real turning point: “The future of bond yields and mortgage rates will be determined as new data on jobs and inflation get released.”
The current mortgage rate is 6.3%, according to Mortgage News Daily, which is of course much higher than the sub-3% rate that homebuyers from the pandemic era remember, although it’s also a far cry from the 8% peak in October 2023.
“The committee’s projections and Chair Jerome Powell’s remarks indicate that this will be the last interest cut for a while,” Zhao wrote. “Given the underlying economic fundamentals of 3% inflation coupled with a weakening—but not recessionary—labor market, the Fed is likely to hold steady in the near future.
“Mortgage rates are unlikely to fall or rise by much,” she continued.
How mortgage rates affect housing affordability
Mortgage rates are just one piece of the housing affordability puzzle. While it may feel as if it’s the major roadblock in the ability to buy a home—especially having a recent memory of the pandemic housing boom—mortgage rates are only one factor.
To put it in perspective, Zillow reported earlier this year not even a 0% mortgage rate would make buying a house affordable in several major U.S. cities.
Let that sink in.
Even without any interest accrued on a loan, homebuying is still out of reach for the typical American. Much of the affordability crisis has to do with home prices, which are more than 50% higher than in 2020. This has locked out new homebuyers from entering the market and current homeowners from selling.
The mortgage rate drop required to make an average home affordable (to about 4.43%) for the typical buyer is “unrealistic,” according to Zillow economic analyst Anushna Prakash.
“It’s unlikely rates will drop to the mid-[4% range] anytime soon,” Arlington, Va.–based real estate agent Philippa Main told Fortune. “And even if they did, housing prices are still at historic highs.” With 11 years of experience, Main is also a licensed mortgage loan officer.
To be sure, some economists see some light at the end of the tunnel for homebuyers plagued by high mortgage rates and home prices.
“For prospective buyers who have been waiting on the sidelines, the housing market is finally starting to listen,” wrote First American chief economist Mark Fleming in an Aug. 29 blog post. First American’s analysis takes into account inflation, and Fleming said: “The price of a house today is not directly comparable to the price of that same house 30 years ago.”
OpenAI, under increasing competitive pressure from Google and Anthropic, has debuted a new AI model, GPT-5.2, that it says beats all existing models by a substantial margin across a wide range of tasks.
The new model, which is being released less than a month after OpenAI debuted its predecessor, GPT-5.1, performed particularly well on a benchmark of complicated professional tasks across a range of “knowledge work”—from law to accounting to finance—as well as on evaluations involving coding and mathematical reasoning, according to data OpenAI released.
Fidji Simo, the former InstaCart CEO who now serves as OpenAI’s CEO of applications, told reporters that the model should not been seen as a direct response to Google’s Gemini 3 Pro AI model, which was released last month. That release prompted OpenAI CEO Sam Altman to issue a “code red,” delaying the rollout of several initiatives in order to focus more staff and computing resources on improving its core product, ChatGPT.
“I would say that [the Code Red] helps with the release of this model, but that’s not the reason it is coming out this week in particular, it has been in the works for a while,” she said.
She said the company had been building GPT-5.2 “for many months.” “We don’t turn around these models in just a week. It’s the result of a lot of work,” she said. The model had been known internally by the code name “Garlic,” according to a story in The Information. The day before the model’s release Altman teased its imminent rollout by posting to social media a video clip of him cooking a dish with a large amount of garlic.
OpenAI executives said that the model had been in the hands of “Alpha customers” who help test its performance for “several weeks”—a time period that would mean the model was completed prior to Altman’s “code red” declaration.
These testers included legal AI startup Harvey, note-taking app Notion, and file-management software company Box, as well as Shopify and Zoom.
OpenAI said these customers found GPT-5.2 demonstrated a “state of the art” ability to use other software tools to complete tasks, as well as excelling at writing and debugging code.
Coding has become one of the most competitive use cases for AI model deployment within companies. Although OpenAI had an early lead in the space, Anthropic’s Claude model has proved especially popular among enterprises, exceeding OpenAI’s marketshare according to some figures. OpenAI is no doubt hoping to convince customers to turn back to its models for coding with GPT-5.2.
Simo said the “Code Red” was helping OpenAI focus on improving ChatGPT. “Code Red is really a signal to the company that we want to marshal resources in one particular area, and that’s a way to really define priorities and define things that can be deprioritized,” she said. “So we have had an increase in resources focused on ChatGPT in general.”
The company also said its new model is better than the company’s earlier ones at providing “safe completions”—which it defines as providing users with helpful answers while not saying things that might contribute to or worsen mental health crises.
“On the safety side, as you saw through the benchmarks, we are improving on pretty much every dimension of safety, whether that’s self harm, whether that’s different types of mental health, whether that’s emotional reliance,” Simo said. “We’re very proud of the work that we’re doing here. It is a top priority for us, and we only release models when we’re confident that the safety protocols have been followed, and we feel proud of our work.”
The release of the new model came on the same day a new lawsuit was filed against the company alleging that ChatGPT’s interactions with a psychologically troubled user had contributed to a murder-suicide in Connecticut. The company also faces several other lawsuits alleging ChatGPT contributed to people’s suicides. The company called the Connecticut murder-suicide “incredibly heartbreaking” and said it is continuing to improve “ChatGPT’s training to recognize and respond to signs of mental or emotional distress, de-escalate conversations and guide people toward real-world support.”
GPT-5.2 showed a large jump in performance across several benchmark tests of interest to enterprise customers. It met or exceeded human expert performance on a wide range of difficult professional tasks, as measured by OpenAI’s GDPval benchmark, 70.9% of the time. That compares to just 38.8% of the time for GPT-5, a model that OpenAI released in August; 59.6% for Anthropic’s Claude Opus 4.5; and 53.3% for Google’s Gemini 3 Pro.
On the software development benchmark, SWE-Bench Pro, GPT-5.2 scored 55.6%, which was almost 5 percentage points better than its predecessor, GPT-5.1, and more than 12% better than Gemini 3 Pro.
OpenAI’s Aidan Clark, vice president of research (training), declined to answer questions about exactly what training methods had been used to upgrade GPT-5.2’s performance, although he said that the company had made improvements across the board, including in pretraining, the initial step in creating an AI model.
When Google released its Gemini 3 Pro model last month, its researchers also said the company had made improvements in pretraining as well as post-training. This surprised some in the field who believed that AI companies had largely exhausted the ability to wring substantial improvements out of the pretraining stage of model building, and it was speculated that OpenAI may have been caught off guard by Google’s progress in this area.
Politics8 years ago