Two of enterprise software's biggest names reported earnings on the same evening — Thursday, September 10, 2026, after the market close — and delivered two very different lessons for investors watching the AI trade unfold.
Oracle surged. Adobe fell. Both beat estimates. The divergence reveals something important about how the market is now separating winners and losers inside the software sector as AI hype gives way to the harder work of measuring actual revenue acceleration.
Oracle: The Cloud Infrastructure Breakout
Oracle Corporation (NYSE: ORCL) has spent years as an afterthought in the cloud computing conversation, perpetually overshadowed by Amazon Web Services, Microsoft Azure, and Google Cloud. That perception is being revised at an accelerating pace.
For its fiscal first quarter 2027, ended August 31, 2026, Oracle reported total revenue of $19.35 billion — up 30% year over year and above the $19.14 billion Wall Street consensus. Adjusted earnings per share came in at $1.92, beating the $1.74 estimate by 10.3%. But the headline number that moved the stock was the cloud infrastructure line: revenue from Oracle's IaaS business surged 121% year over year to $7.39 billion, marking the ninth consecutive quarter of accelerating infrastructure growth. CPU and GPU infrastructure revenue alone reached $6.50 billion, up 151% year over year, as Oracle's AI-optimized data centers ran at 97.9% GPU utilization with more than 300,000 GPUs deployed in the quarter.
Oracle also disclosed a remaining performance obligation — essentially contracted future revenue — of $664.00 billion, a figure that makes the company's current revenue run rate look almost quaint by comparison. For the full fiscal year 2027, management raised guidance to at least $90.00 billion in revenue and $8.10 in adjusted EPS.
The stock jumped more than 5% in after-hours trading and was trending approximately 5.43% higher when markets opened on September 11, 2026. As of midday September 11, ORCL is trading at $151.59, down a modest 0.88% as the broader market absorbs the CPI data released the same morning. The 52-week range runs from $114.50 to $329.50 — a wide band that reflects the stock's violent swings during the AI spending boom and bust. Oracle sits 32.4% above its 52-week low but still 54.0% below its all-time high, giving it significant room to reclaim lost ground if the cloud infrastructure growth trajectory holds. The market cap stands at $436.70 billion.
Analyst reaction was swift and broadly bullish. On September 11, UBS analyst Karl Keirstead raised his price target to $250.00 (from $245.00), Barclays' Raimo Lenschow maintained an Overweight with a $252.00 target, and Guggenheim's John DiFucci reiterated a Buy with the Street's most aggressive target at $400.00. The consensus average target sits at $247.83, implying 63.5% upside from the current price.
Adobe: Strong Quarter, Complicated Story
Adobe Inc. (NASDAQ: ADBE) delivered what would normally be a well-received quarter. Record Q3 fiscal 2026 revenue of $6.76 billion — up 13% year over year — beat the $6.69 billion consensus. Adjusted EPS of $6.13 topped the $6.08 estimate. AI-first annualized recurring revenue exceeded $650.00 million, growing more than 150% year over year. Total monthly active users across all products surpassed 1 billion. Free cash flow for the quarter hit a Q3 record of $2.52 billion.
The stock fell anyway.
Three things weighed on sentiment. First, Q4 fiscal 2026 revenue guidance of $6.80 billion to $6.85 billion came in with a midpoint of $6.825 billion — just below the $6.85 billion analyst consensus. In a market that has been rewarding companies that raise guidance by meaningful amounts (see: Oracle above), coming in a fraction light is enough to trigger selling. Second, CEO Shantanu Narayen announced on September 10 that he will hand over the chief executive role to Anil Chakravarthy — president of Adobe's Customer Experience Orchestration business — effective December 1, 2026. Narayen, who has led Adobe for more than 18 years and oversaw its transformation into a subscription-first digital media powerhouse, will move to executive chair. Leadership transitions at Adobe have been a recurring source of investor anxiety in 2026; the stock fell 7% when Narayen first signaled his departure in March. Third, investors are still waiting for concrete evidence that Adobe's massive AI user base — including 100 million-plus creative freemium users growing more than 70% year over year — converts into paid subscription revenue at scale.
As of September 11, 2026, ADBE is trading at $253.24, up 1.77% on the day as the stock bounces modestly from a tough week. The 52-week range spans $190.12 to $370.86 — meaning Adobe sits 31.7% below its all-time high and 33.2% above its 52-week low, reflecting how much value has been eroded since the stock's AI-driven peak. The market cap stands at $100.70 billion and the P/E ratio is 14.13 — a remarkably compressed valuation for a company growing revenue at 13% with 40%-plus free cash flow margins.
Analyst consensus is Hold, with an average target of $264.17 — just 4.3% above the current price. JP Morgan's Arti Vula is the most optimistic with a $315.00 target. Citigroup's Tyler Radke cut his target to $250.00 from $301.00 on September 11, matching the broad Street caution.
What to Watch
The contrast between Oracle and Adobe is a useful map for how to think about the AI trade inside software right now. The market is rewarding infrastructure capacity and backlog visibility — Oracle's $664.00 billion RPO is essentially a multi-year guarantee of demand. It is being more skeptical of user growth and AI features that have not yet fully flowed through to subscription revenue.
For Adobe, the pivotal question is whether the December 1 CEO handoff brings a fresh strategy that resolves the freemium monetization debate, or whether the transition period creates additional uncertainty. For Oracle, the question is whether it can physically build and operate data centers fast enough to fulfill $664.00 billion in obligations. Both are solvable problems — but the market is currently much more confident in Oracle's answer.
Found this helpful? Share it with others.