Suspect Charged with Attempted Murder in Molotov Cocktail Attack on OpenAI CEO Sam Altman's Home

At 3:40 a.m. on Friday, April 10, 2026, a surveillance camera outside Sam Altman's Russian Hill residence in San Francisco captured a figure in dark clothing hurling a bottle with a burning rag at the CEO's metal gate [1]. The improvised incendiary device — a Molotov cocktail — sparked a small fire that Altman's private security team quickly extinguished [3]. No one was physically injured. But the attack, and what investigators found on the suspect afterward, signaled something more alarming than a single act of arson: it was, prosecutors allege, the opening move of a premeditated campaign to kill the leaders of the artificial intelligence industry.

The suspect, 20-year-old Daniel Moreno-Gama of Spring, Texas, was arrested roughly an hour later outside OpenAI's Mission Bay headquarters, where he was allegedly attempting to smash the building's glass doors with a chair while declaring his intent to burn it down [2]. Officers found kerosene, additional incendiary devices, a blue lighter, and a handwritten manifesto in his backpack [9]. On April 13, the San Francisco District Attorney charged Moreno-Gama with two counts of attempted murder — one for Altman, one for the security guard present at the residence — along with arson and destructive-device charges [1]. Federal prosecutors added counts of attempted property destruction by explosives (carrying a mandatory minimum of five years and a maximum of 20) and possession of an unregistered firearm (up to 10 years) [10].

The Suspect and the Manifesto

Moreno-Gama was a student at Lone Star College in the Houston area [5]. His family's home in a subdivision near The Woodlands was raided by the FBI on the Monday following the attack [3]. His father's neighbors and college acquaintances described him as quiet and unremarkable, with no prior criminal record [9].

The manifesto, titled "Your Last Warning," was a three-part document that outlined a radical anti-AI ideology [10]. In it, Moreno-Gama allegedly claimed responsibility for the attack, argued that artificial intelligence posed an existential threat to humanity, and called on others to commit similar acts of violence against AI executives and investors [2]. The document included names and home addresses of multiple CEOs, board members, and investors at AI companies [1]. It closed with a letter addressed directly to Altman: "If by some miracle you live, then I would take this as a sign from the divine to redeem yourself…" [10].

On the day of the attack, Moreno-Gama emailed a version of the document to representatives at his former college in Texas [10]. Investigators also discovered that he had checked into a Union Square hotel on April 6, stayed two nights, then returned on April 9 — the night before the attack — suggesting days of planning and surveillance [9].

A 9mm handgun and a computer were later found in his hotel room after housekeeping reported that he never returned to check out [9].

A Second Attack — Unrelated but Alarming

Two days after the firebombing, at approximately 2:56 a.m. on Sunday, April 12, police responded to reports of gunfire near Altman's Russian Hill residence [11]. Surveillance footage showed a Honda sedan stopping near the property at 1:40 a.m., then doubling back. A single shot was fired from the passenger window [6].

Police arrested Amanda Tom, 25, a San Diego State University student who lived nearby, and Muhamad Tarik Hussein, 23, her boyfriend, on charges of negligent firearm discharge [11]. Three additional firearms were seized from their home [11].

Unlike the Moreno-Gama case, there is no established ideological motive. Hussein's father told reporters he had "never heard Hussein speak about Altman or AI" [11]. An OpenAI spokesperson said the company believed the incident was unrelated to the firebombing [12]. Police have not confirmed a connection between the two events [6].

The proximity of two attacks on the same residence within 48 hours, however, has compounded security fears across the technology sector.

Threats Against Executives: A Worsening Trend

The assault on Altman's home did not occur in a vacuum. Documented threats and physical attacks against corporate executives have been accelerating in the United States. According to a February 2026 study by the Security Executive Council, published by ASIS International, reported targeting incidents reached 95 in the first ten months of 2025 alone — nearly triple the 33 recorded in 2023 [13]. The technology sector accounted for 17% of all incidents, tied with the financial industry as the most-targeted category [13].

Source: ASIS International / Security Executive Council

Data as of Feb 1, 2026CSV

The most prominent inflection point was the December 2024 murder of UnitedHealthcare CEO Brian Thompson, who was shot outside a Manhattan hotel by Luigi Mangione, a 26-year-old who framed the killing as a protest against the health insurance industry [14]. That event prompted a surge in corporate security spending: UnitedHealth Group disclosed $1.7 million in executive protection costs in the final weeks of 2024 alone [14]. Allied Universal, a security services firm, reported that requests for executive protection assessments increased 10 to 15 times following the Thompson killing [15].

Source: SEC Filings / Fortune

Data as of Apr 1, 2025CSV

Meta has long spent the most on executive protection of any publicly traded tech company, disclosing $23.4 million for Mark Zuckerberg's security in recent filings [15]. OpenAI's comparable figure — approximately $3.5 million for Altman — is significant but substantially lower, raising questions about whether the company's protective posture matched the threat environment.

"Executives are more vulnerable than ever. Across the country, threats are going up," Kent Moyer, CEO of The World Protection Group, told the SF Standard after the Altman attack [3]. He noted that Altman's multiple properties were easily located through public records, a common vulnerability.

Prior to the Molotov cocktail incident, there had been several other AI-related security events: a shooting at an Indiana official's home accompanied by a "No data centers" note, and an anti-AI activist who chained himself to OpenAI's campus in November 2025 before going missing [3].

The Legal Case: Charges and Sentencing Precedent

Moreno-Gama faces both state and federal charges. Under California Penal Code Section 664/187, attempted murder that is willful, deliberate, and premeditated carries a sentence of 15 years to life in state prison [16]. The presence of an incendiary device adds additional charges under California Penal Code Section 18745, which provides for up to life imprisonment when explosives or destructive devices are used with intent to kill [17].

The federal charges carry their own weight. Acting Attorney General Todd Blanche stated: "Violence cannot be the norm for expressing disagreement, be it with politics or technology or any other matter" [10]. U.S. Attorney Craig Missakian indicated that prosecutors are evaluating whether the evidence supports a domestic terrorism enhancement, which would not add separate charges but could influence sentencing [10].

Comparable cases offer a rough sentencing guide. Mangione, Thompson's killer, was convicted of first-degree murder in New York. Ideologically motivated arson cases in California — including attacks on government buildings and corporate offices — have typically resulted in sentences ranging from 10 to 25 years, depending on whether injuries occurred and the degree of premeditation established at trial.

Principled Dissent vs. Political Violence

The Moreno-Gama case has forced an uncomfortable reckoning in AI policy circles. His manifesto reportedly drew on real concerns — job displacement, existential risk, lack of transparency — that are shared by mainstream researchers, ethicists, and even former OpenAI employees [18].

The timing of the attack amplified this tension. Just days before the firebombing, The New Yorker published a major investigation drawing on more than 100 interviews and 200 pages of internal documents, including memos by former chief scientist Ilya Sutskever alleging that Altman exhibited "a consistent pattern of lying" about safety priorities [19]. OpenAI's former policy lead, Miles Brundage, had previously accused the company of "rewriting its AI safety history" [20]. The company's updated safety framework de-emphasized risks from mass manipulation and disinformation — categories previously classified as critical threats [21].

These are legitimate, documented criticisms aired through institutional channels: journalism, congressional testimony, academic papers, and employee whistleblowing.

Moreno-Gama's manifesto, by contrast, called for the murder of executives and provided target lists with home addresses [10]. Whatever overlap exists in the underlying anxieties about AI, the distinction between advocacy and assassination is categorical, not a matter of degree.

John Orloff, a security consultant at Jensen Hughes, observed that tech executives face a particular species of public anger because "ultimately, the CEO is the person held responsible for decisions made by the company" [3]. But that accountability framework — which has historically operated through regulation, litigation, and market pressure — breaks down when individuals appoint themselves as both judge and executioner.

Several AI ethics researchers cautioned against allowing the attack to discredit substantive criticism. The risk, they argued, is that legitimate calls for regulation or slowdowns get conflated with extremism — providing companies a rhetorical shield against accountability.

OpenAI's Response and the Policy Landscape

Hours after the firebombing, Altman published a blog post that included a rare photo of his family. "The fear and anxiety about AI is justified," he wrote. "We are in the process of witnessing the largest change to society in a long time, and perhaps ever." He urged de-escalation: "While we have that debate, we should de-escalate the rhetoric and tactics and try to have fewer explosions in fewer homes, figuratively and literally" [7].

The attack came days after OpenAI published a 13-page policy document titled "Industrial Policy for the Intelligence Age: Ideas to Keep People First," proposing a public wealth fund, higher capital gains taxes, a robot tax, and a four-day workweek [7]. Scholar Anton Leicht characterized the proposals as seeking "deployment absorption instead of development friction" — a framework designed to smooth public acceptance of rapid AI deployment rather than constrain the pace of development [7].

The broader corporate picture is turbulent. OpenAI's CFO Sarah Friar has reportedly been excluded from investor meetings after raising concerns about the company's readiness for a potential 2026 initial public offering and Altman's spending plans [8]. The internal tension between aggressive growth and financial prudence predates the attack, but the security crisis adds a new variable for investors already evaluating leadership risk.

OpenAI has begun hiring for expanded security roles, including an industrial security lead in San Francisco [3]. The company has not disclosed specific changes to Altman's personal protection arrangements, though the presence of a security guard who was on-site during the attack suggests a pre-existing protective detail.

What Remains Unknown

Several questions remain unanswered. Investigators have not disclosed whether Moreno-Gama had connections to any organized anti-AI network or acted entirely alone. The manifesto's distribution — emailed to his college, carried on his person — suggests a lone actor seeking posthumous influence rather than a coordinated cell, but the FBI raid on his family home in Texas may yield additional evidence [3].

The relationship between the firebombing and the Sunday shooting remains officially unconfirmed as connected. The coincidence of two attacks on the same address within 48 hours is striking, but the available evidence points to independent events — one ideologically driven, the other potentially incidental [6][11].

Whether federal prosecutors pursue domestic terrorism enhancements will depend on evidence gathered from Moreno-Gama's computer, hotel room, and communications. The manifesto's explicit call for others to join a campaign of violence against AI executives meets many definitional thresholds for domestic terrorism under federal law, though that designation carries political weight beyond its legal implications.

For the AI industry, the attacks have crystallized a tension that has been building for years: the gap between the speed at which these companies are transforming the economy and the pace at which society — through its institutions, regulations, and public discourse — is processing what that transformation means. The firebombing did not create that gap, but it burned a hole through any remaining pretense that it could be ignored.