Rackspace Technology: AI Rally Exhausted, Guidance Cut And Negative Equity Make This Hold Territory

Rackspace Technology (NASDAQ: RXT) reports second-quarter earnings on August 11, and the setup is far less exciting than the options-implied volatility suggests. The stock has rallied nearly 400% year-to-date on a string of AI partnership announcements, then crashed 33% in a single day when management admitted the AI pivot is cannibalizing existing revenue. What's left is a company with negative book value, $4 billion in debt, contracting core revenue, and an AI growth story whose first meaningful check does not arrive until 2027. The valuation looks cheap on a revenue multiple, but cheap with good reason is not the same as a bargain. I rate this a Hold.

What actually happened

The timeline matters here. Rackspace stock traded as low as $0.39 in early 2026. A strategic partnership with Palantir in February triggered a significant rally. A memorandum of understanding with AMD in May pushed the stock up further. By mid-July, the shares had hit $6.58 and were up roughly 1,600% from their low. The market was pricing in an AI infrastructure turnaround.

Then on July 9, management said the turnaround requires stepping away from the business it has today. Rackspace cut its full-year 2026 revenue guidance by $150 million - from $2.60-2.70 billion down to $2.45-2.55 billion - and reduced adjusted EBITDA (earnings before interest, taxes, depreciation, and amortization) guidance by $20 million, to $285-295 million from $305-315 million. The company simultaneously announced a $250 million at-the-market equity offering, which dilutes existing shareholders, to fund GPU purchases for its AI compute buildout.

The stock fell 33% the next day. It has since recovered to around $4.84, up roughly 27% over the past five days, buoyed by a new Palantir expansion announcement on July 31. But the recovery has been from a deeply beaten-down price after management effectively admitted its legacy cloud business is in structural decline.

The core business is shrinking

Looking past the AI narrative, the operating fundamentals are weak. Q1 2026 revenue was $678 million, up just 2% year-over-year. But that modest growth masks a divergence: private cloud revenue fell 6% to $235 million, while public cloud rose 7% to $443 million. Full-year 2025 revenue of $2.69 billion was down 2% from the prior year.

The July 9 guidance cut makes the trajectory clearer. Management expects Q2 revenue of $641-649 million, below the consensus estimate of $657 million. Public cloud is now expected at $399-403 million for the quarter, down sharply from the $443 million reported in Q1. Private cloud is expected at $242-246 million. Management is deliberately walking away from low-margin infrastructure resale and basic colocation hosting - revenue it knows how to generate today - to free capacity for AI deployments that won't meaningfully contribute until next year.

That is a genuine strategic trade-off if the AI side executes. It is also a massive near-term risk. The company is burning through its proven revenue base while the AI revenue bridge remains under construction.

The AI thesis needs proof, and the clock is unforgiving

Rackspace's growth story now rests on its AI infrastructure plan. Management says the first AMD-backed deployment - nearly 2 megawatts of compute with roughly $75 million in capex - is targeted for completion by year-end. The plan scales to 15 megawatts by end-2027 and 30 megawatts by end-2028. Each deployed megawatt is expected to generate $15-20 million in annual revenue with EBITDA margins above 50%.

The math is plausible. At full deployment, 30 megawatts at $15-20 million each could produce $450-600 million in annual revenue. On a current revenue base of roughly $2.5 billion, that represents a 20-25% lift by 2028. The margin profile - above 50% EBITDA - is dramatically richer than the company's current blended 10.4% EBITDA margin and would be transformative if achieved.

But there are three problems. First, the timeline pushes meaningful revenue contribution into 2027 at the earliest, meaning the stock has to sustain through a year of shrinking top-line numbers and dilution before the AI story shows up on an income statement. Second, the $250 million ATM offering is entirely primary shares - new equity sold to fund capex - which dilutes per-share metrics before the revenue benefits arrive. CFO Mark Marino acknowledged as much on the guidance call. Third, the company has no track record of building GPU infrastructure at this scale. Partnerships with AMD, Palantir, VMware, and Rubrik are promising, but they are commitments, not revenue.

The balance sheet is a red flag

The numbers on the balance sheet are what make me cautious enough to hold rather than buy into the recent bounce. Rackspace carries $3.98 billion in total debt with only $94 million in cash. Net debt sits at $2.66 billion. More striking, total equity is negative $1.22 billion, giving the company a negative book value per share. The current ratio (current assets divided by current liabilities, a measure of short-term liquidity) is just 68%, meaning the company cannot cover its near-term obligations from its near-term assets without access to its revolving credit facility.

The company also has a securities class action lawsuit pending, filed under the caption Morgan-Reed v. Rackspace Technology, Inc., in the Southern District of New York. The complaint alleges Rackspace misled investors about its AI efforts and their impact on financial performance, covering the class period from May 7 to July 8, 2026 - precisely the window between the AMD announcement and the guidance cut. The lead plaintiff deadline is September 28, 2026. The lawsuit does not change the business case directly, but it signals that investors who bought the AI rally at $5-6 per share believe management oversold what it could deliver.

Valuation is cheap, but cheap with a reason

At roughly $4.84, Rackspace has a market capitalization of $1.23 billion and an enterprise value (market cap plus net debt) of $3.89 billion. The stock trades at 0.45x trailing revenue and 1.44x EV/sales - numbers that look inexpensive on their face. For comparison, a mature infrastructure name like NetApp trades at 5.3x revenue with positive earnings and a 1.1% dividend yield.

But comparing Rackspace to a profitable infrastructure company misses the point. The negative P/E, negative book value, negative ROIC (-5.9%), contracting revenue, $4 billion debt load, and dilutive capital raise explain why the multiple is so compressed. Cheap on a revenue multiple is not a buying signal when equity is negative and the growth story is pre-revenue. The valuation has already absorbed a lot of bad news from the July 9 collapse, and the recent 27% five-day rebound has partially unwound that reset. The risk/reward from here is asymmetric in the wrong direction unless the August earnings surprise meaningfully to the upside.

What to watch on August 11

The earnings report itself will show whether Q2 revenue fell as much as the pre-announcement guidance suggested. The company guided to $641-649 million versus a consensus of roughly $657 million. A beat on that range would be the first positive data point in months. More important will be management commentary on the timing of the first AMD deployment, whether the $250 million ATM is fully utilized or partially drawn, and any update to the 2027 AI revenue timeline. If management can show the first megawatt is on track for Q4 delivery and the ATM raise was smaller than the maximum authorized, the stock could find renewed momentum. If the deployment slips or the dilution is heavier than expected, the current level looks like a bounce, not a bottom.

Rating: Hold

Rackspace is not a stock to own right now. The AI partnerships are real, the margin math on AI infrastructure is attractive, and the valuation looks cheap on a revenue basis. But the balance sheet is stretched, equity is negative, revenue is contracting, the dilutive capital raise has already been announced, and the first meaningful AI revenue contribution is at least a year away. The recent rebound from the July 9 low has restored some upside from the post-crash trough, but the risk/reward does not yet favor a buy. Wait for the August 11 report to see whether execution is catching up to the narrative, or for the stock to retest its recent lows before committing capital.

The metrics that would change my view: a Q2 revenue beat, confirmation that the first AMD deployment is on track for end-of-year delivery, and a smaller-than-expected ATM draw. The metrics that would confirm the hold or push to a sell: revenue missing the guided range, deployment delays pushed into 2027, or full utilization of the $250 million ATM.