Rumble Inc.

Revenue in Q2 2026 was $40.4 million, up 61% from $25.1 million in Q2 2025. Cost of services was $30.6 million versus $26.5 million a year ago; general and administrative expense was $16.3 million versus $11.7 million; R&D was $6.8 million versus $4.8 million; and sales and marketing was $10.4 million versus $7.9 million. Adjusted EBITDA loss was $16.6 million, improved from a $20.5 million loss in Q2 2025. Net loss was $80.3 million, or $79.1 million attributable to RUM Group, compared with a $30.2 million net loss a year ago, mainly due to $28.3 million of acquisition-related transaction costs and higher depreciation and amortization. Liquidity ended at $220.5 million, including $203.3 million of cash and cash equivalents and 293.14 Bitcoin valued at about $17.2 million. For Q3 2026, revenue is expected to be between $87 million and $93 million.

Chris Pavlovski described the quarter as the point where Rumble meaningfully entered the cloud and agentic AI era, saying the Northern Data acquisition was the transformation he had previewed last quarter. He argued that Quake AI will become the company’s financial engine, citing more than 85% GPU utilization, the Together AI agreement, and a pipeline tied to 250 megawatts of targeted power for 2027. His tone was strongly optimistic and strategic, with emphasis on AI infrastructure, video data, and a future robotics-era monetization opportunity.

Mike Masci emphasized that this was a landmark and transformational quarter, highlighting record revenue and the first formal revenue guidance. He walked through the cost structure: cost of services at $30.6 million, G&A at $16.3 million, R&D at $6.8 million, and sales and marketing at $10.4 million, with the net loss largely affected by $28.3 million of acquisition-related transaction costs and higher noncash depreciation and amortization. He also noted total liquidity of $220.5 million and said the company spent about $47 million in investing activities tied to IT CapEx for AI compute deals; he added that future reporting will shift to segment disclosure for Rumble Video and Quake AI.

Analysts focused on whether AI capacity will eventually catch up to demand, and Pavlovski said he does not see that happening in the next 1 to 2 years because demand for inference and agentic AI should keep rising. Questions also centered on M&A strategy, equipment sourcing and timing for new data centers, and whether the company might use a powered-shell model; Masci said the company signs deals only when it has strong line of sight to equipment and capacity, and that AI compute-as-a-service offers much higher monetization than leasing power and shell. On advertising, Masci said roughly $4.8 million of Q2 revenue was related to Tether.