Dell Technologies sees more than $1 trillion in AI infrastructure opportunity through 2030 as inference (running trained models to generate outputs) and enterprise agentic AI (AI that performs tasks autonomously) expand data-center demand. Chief Operating Officer Jeff Clarke outlined the outlook during Dell’s fiscal 2027 second-quarter earnings call. Dell booked a record $60.9 billion in AI orders during the quarter and finished with a $95 billion AI backlog. It booked $131.7 billion in orders over the past four quarters, while its five-quarter pipeline grew sequentially. Dell forecasts inference-driven token demand will increase 87 times to 3,600 quadrillion tokens by 2030, while training demand rises fivefold to 850 zettaflops, a measure of computing performance. Enterprise agentic AI could become the largest workload by 2028, and AI could represent 75% of total data-center demand by 2030, requiring an additional 200 gigawatts of power. Dell says about half of that incremental demand will come from neocloud providers, sovereign customers and enterprises, where it already has a significant presence. The company also reported adjusted diluted earnings per share of $7.04 and revenue of $46.97 billion, beating consensus estimates of $4.91 and $44.95 billion, respectively. For the third quarter, Dell forecast adjusted EPS of $6.50 and revenue of $49 billion, above analysts’ estimates of $4.49 and $41.43 billion. Shares closed Tuesday at $425.00, down 6.98%, before rising 8.01% to $459.03 in after-hours trading, according to Benzinga Pro. Benzinga Edge Stock Rankings placed Dell in the 99th percentile for Momentum, with positive trends across short-, medium- and long-term periods.