According to current and former officials, the Biden administration does not have concrete plans to notify the public about deepfakes or other misleading material during the 2024 election unless it is obviously coming from a foreign player and poses a sufficiently serious threat.
While both government and non-government cyber experts anticipate a deluge of misinformation and deepfakes during this year’s election campaign, FBI and DHS officials continue to worry that if they get involved, they will be accused of trying to sway the results in favor of President Joe Biden‘s reelection.
The Biden administration has been pushed by lawmakers of both parties to adopt a more forceful posture.
At a meeting last month, cybersecurity and intelligence officials were addressed by Sen. Angus King, an independent from Maine who caucuses with the Democrats. King expressed his fear to them, saying, “I worry that you might be too preoccupied with looking biased, which would prevent you from acting appropriately.”
Sen. Marco Rubio of Florida, a Republican, questioned the government’s response to a deepfake video. Who is in charge of reacting in the event that this occurs? “Have we considered our course of action in case one of these scenarios materializes?” he inquired. “We would like you to be aware that the video is not authentic. Who would be responsible for that?
Reluctant to call out disinformation with a domestic origin are federal law enforcement agencies, especially the FBI, according to a senior U.S. official familiar with government considerations.
According to the official, the FBI would look into potential election law infractions but is ill-prepared to comment publicly on misinformation or deepfakes created by Americans.
“The FBI does not work in the truth-finding industry,” the spokesperson declared.
The official stated that during interagency meetings regarding the matter, it is evident that the Biden administration lacks a clear strategy for handling domestic election disinformation, encompassing deepfakes that mimic candidates or fabricated reports about violence or closed polling places that may deter voters from casting their ballots.
The FBI admitted in a statement to NBC News that it is unlikely to quickly identify bogus information, even when it looks into potential criminal breaches involving it.
The release stated, “The FBI can and does investigate claims of Americans disseminating false information meant to impede or deny someone’s right to vote.” “The FBI is treating these allegations seriously, and in order to ascertain whether federal law has been broken, we must conduct logical investigation procedures. It is not possible to finish those investigation processes “at the moment.”
In order to exchange information in real-time, the bureau stated that it will “cooperate closely with state and local election officials.” However, as state election authorities oversee elections, the FBI would defer to them over their individual strategies for dealing with misinformation at the time.
State and local election agencies were best suited to alert the public about misleading information propagated by fellow citizens, according to a senior official at the Cybersecurity and Infrastructure Security Agency (CISA), the federal agency in charge of safeguarding election infrastructure. However, the official did not completely rule out the possibility that the agency would issue a public warning if necessary.
“I won’t say that we wouldn’t discuss things in public. I wouldn’t state that in a definitive sense. No, I believe it just depends,” the representative stated.
Is there anything about this that is unique to a state or jurisdiction? Is there more than one state where this is occurring? Does this have an actual effect on the infrastructure used for elections? stated the official.
According to the source, CISA has concentrated on educating the public and educating state and municipal election officials about the strategies used in disinformation campaigns.
“We at CISA have not stopped giving this top priority since we view it as a threat vector for this election cycle” the person stated.
The present federal strategy, according to Robert Weissman, head of Public Citizen, a democracy advocacy organization that has been pushing states to outlaw political deepfakes, is sure to destabilize affairs.
According to him, the greatest concern is a late-breaking deepfake that might harm a candidate’s reputation and affect the result of an election. Government organizations, ranging from federal authorities to county electoral boards, currently have no strategies to address such a development, he said.
According to Weissman, “Even if a tool is immoral, political operatives are likely to utilize it if they have access to a legal one” “Any other expectation than a tsunami of deepfakes is ridiculous.”
Disinformation intended to prevent voters from casting their ballots is forbidden under federal law, but thirty state and federal laws do not forbid deepfakes that misrepresent a candidate’s conduct.
Election officials nationwide have received a warning from the Department of Homeland Security that generative artificial intelligence may enable malicious actors, whether domestic or foreign, to pose as election officials and disseminate misleading information, as has been observed in other nations recently.
A senior federal official in cybersecurity admitted that artificial intelligence (AI)-generated fake videos or audio samples could be dangerous during an election year during a recent meeting with industry executives and nonpartisan watchdog groups. However, they stated that because to the divisive political environment, CISA would not attempt to step in and alert the public.
Intelligence agencies claim to be closely monitoring the false information propagated by foreign adversaries. Recently, officials stated that they are ready to make a public statement about specific disinformation if needed, provided that the source of the false information is unmistakably a foreign actor and the threat is “severe” enough to affect the outcome of the election. However, they haven’t explained just what “severe” entails.
During last month’s Senate Intelligence Committee hearing on the disinformation threat, lawmakers stated that the government needed to develop a more cohesive plan for handling a potentially harmful “deepfake” that could occur during the election campaign.
The chair of the committee, Sen. Mark Warner, D-Va., told NBC News that the federal government must be prepared to act because generative AI poses a “serious and rampant” threat.
“Even as I continue to press tech businesses to take greater action to stop malicious AI content of all kinds, the federal government should have a process in place to alert the public when a real threat arises from a foreign foe.” the congressman stated. “State and federal law enforcement may be in a position to determine, in domestic contexts, whether disinformation related to elections constitutes criminal activity, such as voter suppression.”