Don’t stop at Grok, DICT told as lewd deepfakes spread
Gabriela Rep. Sarah Elago on Saturday said the government’s fragmented response to the growing threat posed by artificial intelligence (AI)-assisted technologies that are used to generate fake, sexually explicit images, is not enough.
In a statement, Elago called on the government to follow in the footsteps of Malaysia and Indonesia, countries that have restricted the use of AI chatbot Grok, as the use of the tool to generate nonconsensual illicit deepfakes of women and minors have spread.
The lawmaker said the Department of Information and Communications Technology’s (DICT) “piecemeal responses are not enough to address the growing threat of artificial intelligence being weaponized against women, LGBTQ+, and even children.” (See related story on this page)
“[Violence] against women has evolved. Abuse now happens not only in homes and public places, but also in digital spaces powered by algorithms and AI tools that profit from women’s objectification,” she said in a statement. “It is time for our laws to catch up with the fast pace of technology especially as it also brings forward new forms of harassment and abuse.”
The DICT on Friday ordered the takedown of Grok, a tool on the social media platform X (formerly Twitter), due to several lapses in regulating content. It said Grok’s affiliates and mother company, xAI, have reached out to local authorities to propose safeguards.
Not enough
The takedown move invoked Republic Act No. 10175, or the Cybercrime Prevention Act, to protect the public from Grok’s ability to “manipulate content, produce sexually explicit materials, and generate deepfakes of real individuals without their consent.”
Elago, however, said RA 10175 was “not enough to sufficiently recognize AI-enabled sexual abuse and exploitation, including nonconsensual deepfake manipulation, as forms of gender-based violence.”
To address the gaps, Elago earlier filed House Bill 4779, which seeks to expand and strengthen Republic Act No. 9262, or the Anti-Violence Against Women and Their Children (VAWC) Act of 2004, to expressly cover online and technology-facilitated forms of abuse.
It was not enough to get reassurances from big tech companies, Elago said, as she called on President Marcos administration to certify the bill as urgent.
In her explanatory note for the bill, Elago said RA 9262 does not “explicitly acknowledge” the evolving form of electronic VAW, which she defined as “any acts or omissions that involve the use or exploitation of information and communications technology which causes or is likely to cause mental, emotional, or psychological distress or suffering to the victim.”
One of seven such acts that she listed in the bill is the use of the victim’s picture, video, voice, name or any other aspect of identities in any video game, phone application, program, and the likes, which puts the victim in a bad light.
Safeguards
On Friday, Communications Technology Secretary Henry Aguda said the National Telecommunications Commission (NTC) had already moved to block Grok, but its affiliates and parent company, xAI, reached out to arrange a meeting with the DITC, NTC, and the Cybercrime Investigation and Coordinating Center to propose safeguards.
Developed by Musk’s xAI, Grok allows users to generate and edit images. However, it has recently been misused to alter photos of real people to show them in sexually explicit or revealing clothing.
Aguda said the DICT generally does not favor blocking digital platforms, but stressed that other government agencies are mandated to monitor and address harmful online content, especially those involving children.
‘Geoblocking’
He said the DICT was already in talks with xAI and “rushing to be the first” to lift the ban once sufficient safeguards are in place.
In an announcement made on the platform, X said it has “implemented technological measures to prevent the Grok account from allowing the editing of images of real people in revealing clothing.”
It would use “geoblocking” to restrict such requests in jurisdictions where the content is illegal, it added.
Malaysia and Indonesia were the first two countries to block Grok due to concerns over its misuse. Authorities in Australia and the United Kingdom are also investigating the matter. —WITH REPORTS FROM LOGAN KAL-EL M. ZAPANTA AND PNA
******
Get real-time news updates: inqnews.net/inqviber
