Two Suspects Arrested in Shooting at OpenAI CEO Sam Altman's Home

On the morning of April 10, 2026, a homemade incendiary device struck the exterior gate of OpenAI CEO Sam Altman's San Francisco residence. Two days later, gunfire hit the same property from a passing car. Within 72 hours, three people were in custody, federal prosecutors had invoked domestic terrorism, and the AI industry was confronting a question it had never faced so directly: whether the backlash against artificial intelligence had turned violent.

The two incidents appear unrelated. But their rapid succession at the home of the most prominent figure in AI has forced a reckoning — among law enforcement, the tech industry, AI critics, and policymakers — about the security of private-sector leaders whose decisions increasingly carry national-security weight.

The Firebombing: April 10

At approximately 4 a.m. on Friday, April 10, Daniel Alejandro Moreno-Gama, a 20-year-old from Spring, Texas, threw a Molotov cocktail at the gate of Altman's home in San Francisco's Russian Hill neighborhood [1]. The device set the exterior gate alight but caused no structural damage and no injuries. A private security guard was on duty at the property at the time [2].

Less than an hour later, Moreno-Gama allegedly walked roughly three miles to OpenAI's Mission Bay headquarters, where he struck the glass doors with a chair and told security personnel he had come "to burn it down and kill anyone inside" [3]. He was arrested outside the building that morning.

The attack was not spontaneous. According to the federal criminal complaint, Moreno-Gama had checked into a Union Square hotel on April 6 — five days before the attack — and checked in again on April 9 [4]. Security footage captured him leaving his hotel room at 2:34 a.m. on April 10, dressed in black and carrying a backpack, roughly two hours before the firebombing [4].

Police recovered incendiary devices, a jug of kerosene, a lighter, a 9mm handgun, and a laptop from his hotel room [4]. Most critically, investigators found a three-part document on Moreno-Gama's person at the time of his arrest.

The Manifesto and the Hit List

The document, described in the federal complaint, laid out views "opposed to Artificial Intelligence and the executives of various AI companies" [3]. The first section, titled "Your Last Warning," advocated for the killing of CEOs of AI companies and their investors [5]. A second section discussed AI's "risk to humanity" and what Moreno-Gama described as "our impending extinction" [6]. The third was 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" [4].

The document also contained a list of names and addresses of board members and chief executive officers of AI companies, along with their investors [3]. Federal authorities have not publicly disclosed which individuals or companies were named beyond Altman and OpenAI.

The Shooting: April 12

Two days after the firebombing, at approximately 1:40 a.m. on Sunday, April 12, a Honda sedan stopped near Altman's property and a shot was fired from the passenger window [7]. Private security personnel at the compound reported hearing the gunshot before the vehicle sped away. The car's license plate was captured on surveillance footage [7].

The San Francisco Police Department traced the vehicle to Amanda Tom, 25, a student at San Diego State University who was previously licensed as a security guard [8]. Tom and her partner, Muhamad Tarik Hussein, 23, lived in a Taylor Street building just blocks from Altman's home, in a property owned by a trust connected to Tom's family [8].

Officers arrested Tom and Hussein without incident at their residence and seized three firearms from the home [7]. Both were booked on suspicion of negligent discharge of a firearm [9].

Hussein's prior criminal record is minor: a 2021 misdemeanor for selling tobacco to someone posing as a minor, which was dismissed after a diversion program, and a 2022 infraction for possessing marijuana while under 21 [8]. His father, reached in Spring, Texas, told reporters he was surprised by the arrest and that his son had never spoken about Altman or AI [8].

Investigators have said there is no evidence connecting the shooting to the firebombing [9]. One detail has drawn scrutiny: Moreno-Gama listed an address in Spring, Texas — the same city where Hussein's father resides — but police have not confirmed any link between the suspects [8].

Charges and the Domestic Terrorism Question

The legal consequences for the two incidents diverge sharply.

Moreno-Gama faces both state and federal charges. At the state level, the San Francisco District Attorney filed two counts of attempted murder — one for Altman and one for the security guard present during the attack — along with attempted arson, carrying potential penalties of 19 years to life in prison [6]. Federal prosecutors added charges of attempted damage and destruction of property by means of explosives (up to 20 years) and possession of an unregistered firearm (up to 10 years) [6].

At a press conference on April 13, U.S. Attorney Craig Missakian declared that authorities "will treat this as an act of domestic terrorism, and together with our partners, prosecute him to the fullest extent of the law" [6]. FBI Acting Special Agent Matt Cobo added: "This was not spontaneous. This was planned, targeted and extremely serious" [6].

The domestic terrorism label, however, is a classification rather than a standalone charge. Under 18 U.S.C. § 2331(5), domestic terrorism is defined as acts dangerous to human life that violate federal or state criminal law and "appear to be intended to intimidate or coerce a civilian population" or "to influence the policy of a government by intimidation or coercion" [10]. Unlike international terrorism, there is no federal statute that criminalizes domestic terrorism as a distinct offense [11]. Prosecutors must instead charge defendants under other applicable statutes — as they have done here — while using the terrorism designation to shape investigative priorities, sentencing enhancements, and resource allocation.

For Tom and Hussein, the charges remain at the state level: negligent discharge of a firearm, a far less severe offense [9]. Without evidence of ideological motivation or intent to kill, prosecutors have no basis for terrorism-adjacent charges.

Security Around AI Executives

The attacks on Altman's home have intensified a conversation about executive protection that was already accelerating before April 2026.

The December 2024 murder of UnitedHealthcare CEO Brian Thompson in Manhattan marked a turning point for corporate America [12]. In the year that followed, the median disclosed executive-security expenditure among S&P 500 companies climbed from $69,180 in 2023 to $94,276 in 2024, and then to a record $106,530 by mid-2025 — a 54% increase over two years [12]. Nearly a third of S&P 500 CEOs now receive company-funded personal security services [12].

Source: ASIS International / Corporate Proxy Filings

Data as of Dec 1, 2025CSV

The spending gap among top tech executives is stark. Meta disclosed $27 million in security costs for Mark Zuckerberg in 2024, more than Apple ($1.4 million for Tim Cook), Nvidia ($3.5 million for Jensen Huang), Alphabet ($6.8 million for Sundar Pichai), and Tesla ($2.8 million for Elon Musk) combined [13]. Musk's figure understates his actual protection, as he operates his own security firm, Foundation Security, staffed in part by former Army Special Forces personnel [13].

Source: Fortune / SEC Proxy Filings

Data as of Aug 16, 2025CSV

OpenAI, as a private company, does not file proxy statements and has not disclosed Altman's security costs. But the company has signaled institutional recognition of the threat: job postings for an industrial security lead and corporate security leadership positions appeared on OpenAI's website in the weeks surrounding the attacks [14].

Kent Moyer, CEO of The World Protection Group, a firm that provides executive protection services, told the SF Standard: "Executives are more vulnerable than ever. Across the country, threats are going up" [14]. Altman's vulnerability is compounded by his maintaining multiple residences — in San Francisco, Napa, and Hawaii — with addresses that are easily discoverable online [14].

No Precedent for Government Protection of Tech Leaders

The U.S. Secret Service protects a defined set of officials: the president, vice president, their families, former presidents, major presidential candidates, and visiting foreign heads of state. There is no statutory framework for extending Secret Service protection to private-sector executives, regardless of the national-security implications of their work [15].

Congress has not introduced legislation to change this. The Secret Service's recent AI-related efforts have focused inward — embedding AI experts within the agency and recruiting technical talent — rather than expanding its protective mandate to cover tech industry leaders [16].

The Domestic Security Alliance Council, established by the FBI in 2005 at the request of corporate security officers, provides a channel for intelligence-sharing between the FBI, DHS, and private-sector executives [17]. But this partnership is informational, not protective. The physical security of executives like Altman remains a private expense.

AI Critics Condemn the Violence — and Resist Conflation

The attacks placed AI safety advocacy groups in a difficult position. Moreno-Gama's manifesto borrowed the language of AI existential risk — "our impending extinction" — that is also used by credentialed researchers and nonprofit organizations working to regulate AI development.

PauseAI, an advocacy group that campaigns for a moratorium on frontier AI development, issued a statement on April 10 "unequivocally condemning this attack and all forms of violence, intimidation and harassment" [18]. The group disclosed that Moreno-Gama had joined their public Discord server roughly two years earlier and posted 34 messages during his membership. None contained explicit calls to violence, though moderators had flagged one message as ambiguous and issued a warning. He held no official role, attended no events, and was banned from the server after the attack [18].

PauseAI drew a sharp line between the attacker's actions and the broader movement: "Advocating safety, regulation and democratic oversight of powerful technologies is a legitimate and necessary part of public life" [18]. The group argued that "organized peaceful movements actually prevent more dangerous isolated actions" — a framing that positions organized advocacy as a safety valve, not a pipeline to violence.

Anthony Aguirre, president and CEO of the Future of Life Institute, which published the widely cited 2023 open letter calling for a pause on AI training, said in a written statement: "Violence and intimidation of any kind have no place in the conversation about the future of AI" [14].

Stop AI, another advocacy organization, stated: "We seek to protect human life. We do not condone any violence whatsoever" while calling for the AI industry to "stop the development of frontier AI systems in the interest of public safety" [14].

The concern among these groups is that the attacks will be used to delegitimize policy criticism. If opposition to AI development becomes associated with firebombings in the public imagination, the political space for regulation narrows. Several commentators have noted this dynamic without endorsing it — the distinction between a 20-year-old with a Molotov cocktail and a peer-reviewed paper arguing that AI poses catastrophic risk is obvious to specialists but may not survive media coverage that collapses both into "anti-AI sentiment."

Altman's Response and the Rhetoric Debate

Altman published a blog post on April 11 responding to both the attack and a critical New Yorker profile that had appeared days earlier. He posted a photo of his husband and their toddler, writing that he shared it hoping "it might dissuade the next person from throwing a Molotov cocktail at our house" [19].

He noted the attack came shortly after what he called "an incendiary article" about him, and that someone had suggested the profile's publication "at a time of great anxiety about AI" could make things "more dangerous" for him [19]. After the attack, Altman wrote that he was "awake in the middle of the night and pissed, and thinking that I have underestimated the power of words and narratives" [19].

The framing drew criticism from some quarters. Critics noted that linking a journalistic profile to a violent attack — even indirectly — risks chilling legitimate reporting and debate. Altman acknowledged the tension, stating that he welcomes "good-faith criticism and debate" while calling for a 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" [19].

He also acknowledged past mistakes, saying he was "not proud of being conflict-averse," which he said had "caused great pain for me and OpenAI" — a reference to his removal and rapid reinstatement as CEO in November 2023 [19].

The Broader Threat Landscape

The attacks on Altman's home did not occur in isolation. The SF Standard reported that an elected official in Indiana had their home targeted, with authorities discovering a note reading "No data centers" [14]. An anti-AI protester who chained himself to OpenAI's San Francisco campus reportedly disappeared after hinting at violent plans [14].

Public unease about AI is substantial and measurable. Roughly half of U.S. adults surveyed express greater concern than excitement about AI developments, citing job displacement and educational impacts [14]. Communities across the country have organized against AI data center construction, citing strain on electrical grids and water supplies [14]. Protests have targeted AI companies' government contracts, particularly with the Pentagon [14].

None of this constitutes a threat of violence. But it forms the backdrop against which isolated violent acts are interpreted — and potentially exploited — by actors on all sides of the AI debate.

What Remains Unknown

Several questions remain unanswered as of April 14, 2026. Investigators have not disclosed the full list of individuals named in Moreno-Gama's manifesto, nor have they established whether Tom and Hussein had any ideological motivation for the shooting or whether it was a random act of recklessness. The potential connection between Moreno-Gama's Spring, Texas address and Hussein's father's residence in the same city has not been resolved.

OpenAI has not publicly disclosed the total number of threats Altman or the company has received, nor has it detailed its security expenditures. The company's status as a private entity means these figures are unlikely to become public absent a legal proceeding or voluntary disclosure.

Whether Moreno-Gama's case will ultimately be prosecuted under a domestic terrorism framework — with its attendant sentencing enhancements — depends on whether prosecutors can establish that the attack was intended to intimidate a civilian population or influence government policy, rather than target a single individual [10]. That legal determination will shape not only this case but potentially the framework for future threats against technology leaders whose work intersects with questions of national security and public safety.