Okay, Now Exactly Half Of xAI’s Founding Team Has Left The Company

xAI Founders Exodus: Half the Team Has Left

Six of xAI's twelve founding team members have now departed the company, with two high-profile exits occurring within 48 hours this week. Co-founders Yuhuai "Tony" Wu and Jimmy Ba announced their resignations on February 10 and 11, 2026, bringing the total number of founding departures to exactly half the original team. Five of these six exits have occurred within the past year alone, coinciding with SpaceX's recent acquisition of xAI and preparations for a mid-2026 public listing that could value the combined entity at over $1 trillion.

Credit: Jakub Porzycki / NurPhoto / Getty Images

The rapid succession of leadership departures has sparked concern among industry observers about xAI's stability as it races to advance its Grok AI models amid intensifying competition. While both Wu and Ba framed their exits as voluntary transitions to "next chapters," internal sources suggest mounting pressure to deliver accelerated technical milestones under Musk's demanding oversight.

A 48-Hour Leadership Collapse

The exodus accelerated dramatically during a single workweek. Late Monday night, Tony Wu posted a brief but reflective announcement on X: "It's time for my next chapter. It is an era with full possibilities: a small team armed with AIs can move mountains and redefine what's possible." His departure came without elaboration on specific grievances or future plans, maintaining the diplomatic tone common in Silicon Valley exit announcements.

Less than 24 hours later, Jimmy Ba—described as reporting directly to Musk—shared his own farewell message. "Enormous thanks to @elonmusk for bringing us together on this incredible journey. So proud of what the xAI team has done and will continue to stay close as a friend of the team," Ba wrote. His phrasing notably avoided commitments to remain actively involved with xAI's technical direction, signaling a clean break despite the gracious wording.

These back-to-back announcements transformed what might have been viewed as isolated career moves into a visible pattern of attrition at the highest levels of Musk's AI venture.

The Full Roster of Departing Founders

xAI launched in 2023 with an elite twelve-person founding team assembled from top AI research institutions including Google DeepMind, Microsoft Research, and the University of Toronto. The complete list of departed co-founders now includes:

• Kyle Kosic, infrastructure lead who left for OpenAI in mid-2024
• Igor Babuschkin, who departed in August 2025 to launch a venture firm
• Greg Yang, a Microsoft veteran who exited in January 2026 citing health concerns
• Christian Szegedy, whose departure preceded the recent wave
• Yuhuai "Tony" Wu, who announced his exit February 10, 2026
• Jimmy Ba, the most recent departure announced February 11, 2026

The concentration of exits within a single twelve-month window stands in stark contrast to the typical founder retention patterns at well-funded AI labs. Competing organizations like Anthropic and Cohere have maintained near-complete founding team continuity through their growth phases.

Timing Raises Questions About Voluntary Departures

The departures coincide with significant corporate restructuring. SpaceX finalized its acquisition of xAI in early February 2026, folding Musk's AI division into the space company's operations ahead of a planned June IPO. This merger has triggered organizational realignment, with engineering teams reportedly facing intensified pressure to accelerate Grok's development timeline.

Musk himself has suggested some exits may not be entirely voluntary. In comments following earlier departures, the billionaire indicated certain researchers were "pushed" rather than choosing to leave—a characterization that contrasts with the polished farewell posts appearing on X. Internal communications reviewed by industry analysts describe leadership overpromising technical milestones to Musk, creating unsustainable delivery expectations for engineering teams.

The timing also follows reported layoffs within xAI's data annotation division—the team responsible for training Grok's reasoning capabilities—with approximately 500 positions eliminated in late 2025 as the company shifted focus toward specialized AI tutors. These cuts preceded the recent wave of executive departures, suggesting broader organizational instability.

What This Means for Grok's Development Trajectory

xAI's flagship product, the Grok series of large language models, has become increasingly central to Musk's ecosystem. Grok now powers content moderation across X, assists with engineering tasks at Tesla, and provides analytical support for SpaceX mission planning. The loss of founding researchers with deep institutional knowledge of Grok's architecture could complicate upcoming releases.

The company recently unveiled Grok-4 with enhanced agentic coding capabilities, positioning it against OpenAI's o1 and Google's Gemini models. However, industry analysts note that foundational model development requires sustained leadership continuity—particularly during the delicate transition from research prototypes to production-scale systems. The departure of multiple co-founders who architected Grok's core training methodologies may introduce technical debt or knowledge gaps difficult to overcome quickly.

Musk has responded to the attrition by hosting an all-hands meeting with xAI employees, reportedly outlining "interplanetary ambitions" for the AI division and emphasizing 2026 as "the busiest and most consequential year for the future of our species." Whether this vision resonates with remaining engineers remains uncertain as IPO pressures mount.

Broader Implications for AI Talent Retention

The xAI situation reflects intensifying competition for elite AI researchers across the industry. Compensation packages now routinely exceed $5 million annually for top machine learning scientists, with equity stakes in pre-IPO companies serving as major retention tools. Yet financial incentives alone cannot guarantee stability when cultural mismatches emerge between founders and corporate leadership.

Musk's management style—characterized by aggressive deadlines, public criticism of underperforming teams, and frequent strategic pivots—has proven effective at accelerating certain engineering milestones but may conflict with the methodical, consensus-driven approach many AI researchers prefer. The contrast between xAI's attrition rate and the relative stability at labs led by researchers-turned-CEOs like Dario Amodei or Demis Hassabis suggests leadership philosophy significantly impacts talent retention in foundational AI work.

For enterprise customers and developers building on Grok's API, the leadership turnover introduces uncertainty about long-term product roadmaps and support continuity. Companies integrating Grok into mission-critical workflows must now weigh these organizational risks against the model's technical capabilities.

Can xAI Stabilize Before IPO?

With SpaceX's IPO targeted for June 2026, xAI has approximately four months to demonstrate organizational stability to potential investors. The merged entity's valuation—projected between $1.25 and $1.5 trillion—depends partly on convincing markets that xAI can sustain competitive AI development without its founding architects.

Musk has historically weathered executive turnover at Tesla and SpaceX by personally driving technical direction during transition periods. Whether this approach translates effectively to cutting-edge AI research—where breakthroughs often depend on collaborative insight rather than top-down mandates—remains unproven. The next quarterly update on Grok's performance metrics will serve as a critical indicator of whether xAI's engineering momentum has survived its leadership exodus.

For now, exactly half of xAI's founding visionaries have departed. Their collective exit transforms what began as an ambitious AI moonshot into a high-stakes test of whether Musk's operational intensity can compensate for the loss of the very minds who built his AI ambitions from scratch. The answer will determine not only xAI's fate but also whether founder-led continuity remains essential in the race to build artificial general intelligence.