Small print.
Wide open.

Table of Contents

• mandated  safety,  risk-management,  or  evaluation  frameworks,  certifications,  or  approvals,  including 
requirements for independent, third-party review or approval prior to model development, training, or 
release; for example, in September 2026, a state attorney general asked a court, in pending litigation 
alleging  harm  to  minors,  to  enjoin  an  AI  company  from  developing  new  AI  models  without 
independent, third-party guardrails and approval;

•

•

•

•

•

restrictions on data center development or expansion, including permitting, zoning, or limits on energy 
or water usage;

heightened energy efficiency, sustainability, or environmental compliance requirements;

limitations on cross-border deployment, access, or commercialization of AI models, including through 
export controls, national security determinations, or other governmental measures that could restrict 
our ability to sell our models, platform, and applications in certain markets or to certain customers; for 
example, in June 2026, the U.S. Department of Commerce imposed worldwide export restrictions on 
our Fable 5 and Mythos 5 models, which were subsequently lifted after we updated our safeguards, and 
other models may be subject to these or similar export restrictions in the future; for more information, 
see the section titled “—Governmental trade controls, including export and import controls, sanctions, 
customs requirements, and related regimes, could subject us to liability or loss of contracting privileges 
or limit our ability to compete in certain markets.”;

AI-specific taxes, fees, reporting obligations, or other financial burdens; or

outright  bans  or  moratoria  on  activities  material  to  our  business  or  the  businesses  of  our  customers 
and partners.

Because  AI  regulation  is  developing  rapidly  and  in  a  fragmented  manner,  including  as  a  result  of  political 
considerations, public sentiment, and geopolitical dynamics, we will not be able to fully anticipate how existing 
laws  will  be  interpreted  or  enforced,  how  new  legal  frameworks  will  evolve,  or  whether  regulatory  regimes 
across  jurisdictions  will  converge  or  diverge  over  time.  Compliance  with  new  or  changing  requirements  may 
require  substantial  investment  in  research  and  development,  compliance  systems,  operational  processes,  or 
localized  offerings,  and  we  may  need  to  adjust  or  limit  our  activities  in  certain  jurisdictions  to  address 
conflicting or inconsistent legal obligations. In addition, regulatory actions affecting our customers or partners, 
including  restrictions  on  their  ability  to  develop,  deploy,  or  commercialize  AI-enabled  models,  platforms,  or 
applications,  could  indirectly  reduce  demand  for  our  models,  platform,  and  applications.  Any  failure,  or 
perceived  failure,  to  comply  with  applicable  laws  and  regulations,  or  any  significant  increase  in  the  cost  or 
complexity of compliance, could result in litigation, enforcement actions, penalties, reputational harm, delayed 
or reduced adoption of our offerings, or other adverse effects on our business, financial condition, and results of 
operations.

Additionally,  as  AI  systems  become  more  sophisticated,  emerging  debates  regarding  whether  AI  systems 
possess  or  could  develop  forms  of  sentience  or  moral  status  may  lead  to  regulatory  requirements,  public 
pressure, or advocacy efforts that could constrain how we develop, train, deploy, or modify our models, increase 
our  costs,  or  affect  our  reputation,  regardless  of  the  ultimate  resolution  of  these  philosophical  and  scientific 
questions. In addition, given our focus on developing frontier AI, and the way we approach model sentience and 
moral status, we may face heightened scrutiny or pressure in connection with these debates and could elect to 
adopt  policies,  safeguards,  or  limitations  regarding  model  development,  deployment,  or  modification  that  go 
beyond  existing  legal  requirements,  which  could  constrain  our  operations,  increase  costs,  or  limit  the 
commercial potential of our models, platform, and applications. If we are unable to manage the foregoing risks, 
our business, financial condition, and results of operations could be adversely affected. 

Our sales to government entities and highly regulated organizations are subject to challenges 
and risks. 

Our  customers  include  U.S.  and  non-U.S.  government  agencies.  Revenue  from  government  agency  contracts 
represented, in aggregate, approximately 1% of our revenue in each of the periods presented in this prospectus; 
however,  the  risks  described  below,  including  the  governmental  actions  described  in  this  risk  factor,  are  not 
limited  to  revenue  generated  directly  from  government  agencies  and  could  affect  our  relationships  with 

70