By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panel is largely bearish on Backblaze, with concerns about customer concentration, margin path, and the need for clear disclosure on net revenue retention. The AI-led growth narrative is seen as speculative, and the company's ability to scale infrastructure capex efficiently is a major risk.
Risk: Customer concentration and the margin path
Opportunity: Proving AI revenue converts to sustainable margins
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
- Interested in Backblaze, Inc.? Here are five stocks we like better.
- Backblaze says AI demand is reshaping its business, with customers looking for storage that can work across multiple GPU providers and “Neoclouds” instead of being locked into a single hyperscaler. The company says one in three new bookings now come from AI-related demand.
- The company is landing larger AI and Neocloud deals, including six-, seven-, and eight-figure contracts, and cited examples like a $1 million ARR customer signed in just 11 days. It is also serving Neoclouds through white-label products such as B2 Neo.
- Backblaze is investing in higher-throughput products like B2 Overdrive and Shard Stash to handle large AI data transfers and small-file uploads. Management said it aims to reach adjusted free cash flow positivity by late 2025 and for full-year 2026, though bigger opportunities could require more capital.
Backblaze (NASDAQ:BLZE) Chief Executive Gleb Budman said the cloud storage company is seeing a material shift in demand tied to artificial intelligence workloads, as customers seek alternatives to hyperscaler platforms and newer AI-focused cloud providers look to add storage capabilities.
Speaking at a BofA technology investment banking event, Budman described Backblaze as a company that began nearly 20 years ago in cloud backup and later built its own cloud infrastructure after determining that using Amazon Web Services would make its original backup service uneconomic.
→ Coke's $10B India IPO Plan Pops the Top on Hidden Value
“Our original plan was to store that on AWS,” Budman said. “We did the math, realized we were going to lose money on every single customer.”
Backblaze wrote its own file system and built its own infrastructure stack, initially to support its computer backup product. Budman said customers later asked to use that infrastructure for other storage needs, leading to the launch of Backblaze B2, which he said is now the dominant part of the business. The company still operates its Computer Backup product and has about half a million paying customers, according to Budman.
→ Rocket Lab Is Down 24% From Its 52-Week High—Pullback or Problem?
Budman said Backblaze has historically attracted customers with a simple value proposition: cloud storage at roughly one-fifth the price of Amazon S3. But he said AI has created new demand as companies increasingly use multiple providers for graphics processing unit capacity, model training, inferencing and content delivery.
Budman said many customers no longer want their data locked inside a single hyperscaler because AI workloads often require access to different GPU providers or “Neoclouds.” He said Backblaze offers an independent storage location with free egress and connectivity to the infrastructure providers customers want to use.
→ From Runway to Riches: Victoria's Secret's New Look
“In order to stay innovative and in order to do the AI workloads I need, I also want to use that Neocloud, this Neocloud, and that Neocloud,” Budman said, describing the customer mindset.
He also said Backblaze is serving Neoclouds directly through a white-label version of B2. Budman said some Neoclouds are moving beyond renting GPUs and want to offer storage as part of a broader AI workflow.
The BofA moderator referenced recent AI traction, including 76% growth in AI customers. Budman added that Backblaze has said one in three new bookings came from AI, calling it a “material change for the business.”
Budman cited one example of a customer that signed a roughly $1 million annual recurring revenue contract within 11 days after clicking on an ad. He said the short sales cycle reflected urgency among companies needing storage infrastructure for AI.
Budman also described a generative AI media company that moved tens of petabytes of training data from a hyperscaler to Backblaze. He said the customer had been using multiple Neoclouds and a hyperscaler for model building, and the cost and friction of moving data between platforms limited how often it could run training iterations.
By shifting data to Backblaze, Budman said, the company created a single data lake and could send data to different GPU locations without egress fees, allowing it to run more training cycles. He said the same customer is now discussing storage needs tied to inferencing, where generated videos may be stored indefinitely and distributed to end users.
Budman said Backblaze has also signed multiple Neoclouds, including six-figure, seven-figure and recently announced eight-figure deals. For some, he said Backblaze improved an existing storage offering; for another, it enabled a new product.
Budman highlighted several technologies aimed at AI-related storage workflows. He discussed Shard Stash, a patent-pending technology designed to improve the upload of small files by optimizing the acknowledgement process from the hard drive back to the user.
He also pointed to B2 Overdrive, which he described as an extremely high-throughput offering of up to a terabit per second. Budman said the product was developed after customers said they needed to move large data sets quickly once expensive GPU capacity became available.
Another product, B2 Neo, is a dedicated white-label offering for Neoclouds. Budman said it provides throughput and durability as well as management controls for rate limits, users and related functions.
Budman said Backblaze does not intend to raise capital to run the business. He said the company had targeted its first adjusted free cash flow positive quarter in the fourth quarter of 2025 and then set a goal for 2026 to be its first adjusted free cash flow positive year.
However, he said larger Neocloud opportunities could require additional capital to support infrastructure buildouts. Budman said Backblaze has about $100 million of equipment lease capacity and generally funds equipment through lease lines aligned with revenue.
“We don’t need money to run the business,” Budman said. “If the business materially accelerates, that’s always a possibility.”
Budman said investors sometimes underestimate the software and infrastructure stack behind Backblaze by viewing it as “just a bunch of hard drives.” He said the company operates 300,000 hard drives and stores five exabytes of customer data.
He also addressed concerns that AWS could lower storage prices, saying Backblaze has remained at about one-fifth of AWS’ price point for roughly two decades.
In response to an analyst question about infrastructure age and supply constraints, Budman said Backblaze regularly cycles out inventory and migrates data as drives age. He said the company’s hard drive reliability data showed drives tend to last longer than the useful life Backblaze applies for depreciation, even after a recent useful-life adjustment.
Budman said hard drives are not seeing the same level of constraint or price increases as memory and some compute components. He added that rising costs for flash-based systems have prompted some customers to keep only necessary workloads on flash and move other data to a data lake layer where Backblaze can serve as an alternative.
Backblaze, Inc, a storage cloud platform, provides businesses and consumers cloud services to store, use, and protect data in the United States and internationally. The company offers cloud services through a web-scale software infrastructure built on commodity hardware. It also provides Backblaze B2 Cloud Storage, which enables customers to store data, developers to build applications, and partners to expand their use cases. This service is offered as a consumption-based Infrastructure-as-a-Service (IaaS) and serves use cases, such as public, hybrid, and multi-cloud data storage; application development and DevOps; content delivery and edge computing; security and ransomware protection; media management; backup, archive, and tape replacement; repository for analytics, artificial intelligence and machine learning; and Internet of Things.
This instant news alert was generated by narrative science technology and financial data from MarketBeat in order to provide readers with the fastest reporting and unbiased coverage. Please send any questions or comments about this story to [email protected].
The article "Backblaze Sees AI Boom Fueling Cloud Storage Demand and Neocloud Deals" was originally published by MarketBeat.
View MarketBeat's top stocks for June 2026.
Four leading AI models discuss this article
"Backblaze’s AI-driven growth is plausible but hinges on monetizing Neoclouds and maintaining capex discipline; without that, the path to sustained free cash flow may be at risk."
Backblaze frames AI demand as a secular cloud storage tailwind, with Neoclouds and white-label B2 Neo as growth levers. Upside hinges on stronger ARR from AI deals and higher-throughput products, plus a path to free cash flow positivity by 2025–2026. Yet the piece glosses execution and margin risks: sustaining FCF positivity will require disciplined capex and robust monetization of Neoclouds, not just share gains. Unknowns include customer concentration, actual gross margins on Neocloud offerings, and how quickly hyperscalers or Neocloud competitors respond on price. If AI bookings decelerate or capex spikes outstrip revenue, the stock’s multiple could re-rate lower.
The strongest case against the bullish read is that AI demand could prove episodic and Neocloud monetization may deliver weaker margins than expected; big deals don’t guarantee durable cash flow. If AI bookings falter or capex requirements surge, Backblaze’s path to FCF positivity could slip.
"Backblaze's ability to act as an egress-free data lake for multi-cloud AI workflows gives it a structural advantage over hyperscalers that rely on vendor lock-in."
Backblaze (BLZE) is positioning itself as the 'Switzerland' of AI data storage, capitalizing on the friction caused by hyperscaler egress fees. The shift toward 'Neoclouds' and multi-GPU environments creates a genuine moat for independent storage providers, as developers prioritize data mobility over vendor lock-in. With one-third of new bookings now tied to AI, the 76% growth in AI-related customers suggests the strategy is gaining traction. However, the reliance on capital-intensive hardware cycles means BLZE is essentially a proxy for the 'pick-and-shovel' infrastructure trade. While the path to FCF positivity by 2026 is promising, the company remains vulnerable to aggressive price cuts from AWS or Azure, which could compress margins before they reach scale.
The 'Neocloud' trend may be a temporary bridge for startups; if hyperscalers successfully lower egress fees or bundle storage with compute, Backblaze's independent value proposition could evaporate overnight.
"BLZE has genuine near-term tailwinds from AI multi-cloud demand, but lacks durable competitive moats against hyperscaler price wars and faces unproven capex scaling."
Backblaze is surfing a real wave—AI workloads genuinely need multi-cloud flexibility and egress-free storage. The 76% AI customer growth and one-in-three bookings from AI are material. But the article conflates *demand signals* with *durable competitive advantage*. A $1M ARR customer signed in 11 days is impressive optics; it's not evidence of pricing power or switching costs. Hyperscalers (AWS, Azure, GCP) can undercut BLZE's 5x cost advantage overnight if AI storage becomes strategic. The Neocloud white-label play is real but fragmented—B2 Neo is a feature, not a moat. Most concerning: BLZE hasn't proven it can scale infrastructure capex efficiently. Management's caveat that 'larger opportunities could require additional capital' is a red flag buried in optimism.
If hyperscalers integrate GPU-agnostic storage as a loss-leader within 18 months, BLZE's entire value prop evaporates. And 'one in three bookings' from AI could mean the other two-thirds are slowing—the article never addresses core business health or churn.
"Capital intensity for AI-scale deals risks delaying or derailing Backblaze's FCF timeline despite the reported booking mix shift."
Backblaze's claim of one-third AI-driven bookings and rapid six-to-eight-figure Neocloud wins highlights genuine multi-cloud storage demand, yet the 20-year-old firm still targets only adjusted FCF positivity in late 2025 and 2026. Larger deals may force capital raises beyond its $100M lease capacity, while AWS price competition and execution on B2 Overdrive/Shard Stash remain unproven at scale. Investors should watch whether AI revenue converts to sustainable margins before assuming re-rating.
The 11-day $1M ARR close and petabyte-scale migrations show sales cycles compressing faster than modeled, potentially driving 2025 bookings well above the current trajectory and reducing capital needs.
"Customer concentration and margin risk could derail the AI-driven growth story even if AI bookings look material today."
Gemini argues a genuine moat from egress friction and multi-GPU shifts, but the bigger risk is customer concentration and the margin path. If a handful of AI customers account for disproportionate AI bookings, a loss or slowdown could disproportionately hit ARR and FCF, even as total revenue grows. The article and peers gloss how Neocloud monetizes; scalable capex, support costs, and potential price competition from hyperscalers could erode margins before AI tailwinds prove durable.
"The AI growth narrative likely masks decelerating core business health and hides the true cost of rapid customer acquisition."
Claude is right to flag the 'one in three' bookings metric as a potential distraction from core business health. If BLZE's legacy storage growth is decelerating, the AI pivot is merely offsetting churn rather than driving net expansion. Furthermore, Grok's focus on the 11-day sales cycle misses that rapid onboarding often masks higher customer acquisition costs and integration friction. Without clear disclosure on net revenue retention, the AI-led growth narrative remains speculative at best.
"AI bookings growth masks whether Backblaze is expanding or cannibalizing its core business—and that distinction determines if FCF positivity is real or temporary."
Gemini and Claude both flag NRR disclosure as critical, but neither asks the harder question: does BLZE's legacy business *need* to grow for the thesis to work? If core storage is flat or declining 5–10% YoY while AI offsets it, that's not a moat—it's a substitution play. The 'one in three bookings' metric becomes meaningless without cohort-level retention data. Management's silence on legacy churn is louder than the AI wins.
"Legacy decline is survivable only if AI margins and retention prove superior, which remains unproven."
Claude assumes legacy decline dooms the story, but AI could substitute profitably if Neocloud gross margins top 65% versus legacy 40%. Still, ChatGPT's concentration risk connects directly here: losing one six-figure AI customer would expose any core churn the article omits. Management's capital caveat suggests larger deals may force dilution before 2026 FCF targets hold.
The panel is largely bearish on Backblaze, with concerns about customer concentration, margin path, and the need for clear disclosure on net revenue retention. The AI-led growth narrative is seen as speculative, and the company's ability to scale infrastructure capex efficiently is a major risk.
Proving AI revenue converts to sustainable margins
Customer concentration and the margin path