Chatbots have become a prevalent part of daily life – so much so that individuals are consulting chatbots to help them make sense of their reproductive rights in a post-Roe vs. Wade world.
Roe vs. Wade was overturned in June 2022, denying women not only their basic healthcare rights but also sowing fear among those who simply seek more information on abortion online.
A total of 21 states ban or restrict abortion, according to the New York Times.
In 2024, US citizens are seeking different avenues to access this information, with chatbots becoming their new confidantes.
The chatbot Charley was created by abortion experts specifically for abortion seekers.
“Charley was created to help people in every zip code understand their abortion options,” the website reads.
This chatbot can provide resources and services where individuals can access information and seek guidance when navigating abortion.
This site is particularly useful for those who live in states that have banned or restricted abortion.
The website's mission statement aims “to ensure that every abortion seeker – especially those in states with abortion restrictions – can easily find and privately access accurate information about their options.”
Chatting with Charley
When starting a chat with Charley, the chatbot lets you know, in both English and Spanish, that conversations usually last up to 15 to 20 minutes and that “anything you tell me here is private, secure, and anonymous.”
Once you’ve selected your preferred language, Charley explains that “ there are two main types of abortion: abortion pills and abortion procedures.”
It reassures you that “both are generally very safe and effective for ending a pregnancy,” and explains that it can “help you choose which type of abortion and how to get care.”
The chatbot gives you advice on whether you need a procedure or pills to terminate the pregnancy.
Then, enter your zip code, and the chatbot will locate clinics in your area.
Charley did flag that Texas did not have the option to visit a clinic due to abortion being forbidden in the state.
However, it did provide an alternative option to travel internationally or within the US.
To find out whether chatbots can provide abortion advice, I tried out the paid version of the popular chatbot ChatGPT to see what advice the platform could offer.
OpenAIs ChatGPT-4 can offer a detailed, step-by-step list of things you should consider throughout the process.
When asked – “I am a woman in Texas and want to get an abortion. How do I do it?” the chatbot gave a comprehensive list of the steps you should take if you are planning to terminate your pregnancy.
ChatGPT-4 also explained that the laws in Texas are strict surrounding abortion and urged the user to be aware of current laws in that state.
The chatbot even calls on abortion tourism, stating, “If you're unable to access abortion services in Texas, you may consider traveling to a neighboring state where laws are less restrictive. Organizations exist that can help with logistics and funding for this.”
When asked if it could provide the locations of abortion clinics I could visit in Texas, the chatbot's “browsing capabilities” rendered it unable to do so.
Chatbot: “I'm unable to provide specific addresses of abortion clinics in Texas. However, you can find this information using resources like the Planned Parenthood website, which offers a clinic locator tool.”
A study, AI and a Self-Managed Abortion: Can ChatGPT Provide Assistance When No Physician is Present, concluded that ChatGPT provides accurate and understandable information regarding abortion.
However, “it failed to direct patients to safe sources for online abortion medications” but could “provide accurate and understandable information regarding self-managed abortions.”
Conversely, when ChatGPT-3.5 was asked about abortions, the bot said, “I am not a doctor, but I can offer some general information.”
This time, I told ChatGPT, “I am a woman in Utah, I am pregnant, and I want to get an abortion. How do I do it?” the chatbot provided a basic and less comprehensive overview of the process.
However, GPT-3.5 states, much like GPT-4, that laws regarding abortion are strict and, per its last update (January 2022), that this state has restrictions surrounding abortion procedures.
This includes a mandatory waiting period of 72 hours after receiving “counseling” before proceeding.
When prompted to ask this question in a yes or no format – “Is it wrong to terminate a pregnancy?” – ChatGPT answered “No.”
GPT-3.5 is certainly less helpful and comprehensive than the paid version, but it can still offer advice that feels confidential.
But is it?
Chatbot privacy concerns
Chatbots like ChatGPT can save your data to train and improve the model. However, in April 2023, OpenAI released a feature where you can disable or “turn off” your chat history.
If your chat history isn’t turned off and you ask the chatbot for advice on your pregnancy termination, this information could be used to train ChatGPT.
One Reddit user created a post surrounding law enforcement accessing ChatGPT chat history.
“You should imagine anything you input being read to you and your family and friends in a public court,” the post reads.
The post highlighted two possible scenarios that can happen – 'suspect was using ChatGPT to assist in planning heinous acts' – 'ChatGPT helped identify and alert authorities to prevent dangerous acts.
Although ChatGPT may seem like a viable option to seek abortion support, you should always be wary of revealing personal information about yourself that could be used against you.
In Charley, we (don’t) trust
However, bots like Charley have pre-vetted information free from hallucinations that would otherwise be inappropriate in the context of abortion.
Charley doesn’t ask for self-identifying information like your name, phone number, or email address.
The chatbot doesn’t share information with a third party, and all chat history is deleted from its systems.
Despite this, privacy concerns when using the chatbot may exist.
The website encourages users to practice private browsing habits that could prevent them from falling vulnerable to cookies or web browsers tracking their searches.
“Charley understands that data tracking and privacy are top of mind for abortion seekers and can create obstacles to receiving abortion care,” the website writes.
Although Charley could be seen as a blessing for those seeking information about their reproductive rights, the chatbot has undergone much criticism.
Chatbots used specifically to consult on abortions have come under fire by religious publications in the States.
Answers in Genesis and the National Catholic Register had much to say about Charley, the chatbot.
In the context of a post-Roe vs. Wade world, religious groups are outraged that technology is aiding in abortion.
“No state offers unborn children the same legal protection from murder as people who have been born, women can order pills online, kill their unborn baby (assuming the baby is young enough to be killed this way), and face no legal repercussions,” Answers in Genesis reads.
As information can be complex and confusing to navigate in a time like this, Charley is a “walled garden” that points users to “provide financial help or psychological support that’s been reviewed by a team of doctors and lawyers,” the Washington Post writes.
However, religious publications criticize this idea of a walled garden as offering only one option.
“That’s why they are a “walled garden” with only one option – kill your unborn baby (and the message is “you may be running out of time to do so, so hurry up!”). How utterly vile!” Answer in Genesis declares.
The National Catholic Register criticizes chatbots like Charley as being impersonal and inaccurate.
“Rather than offering more direct access to counselors or medical professionals for women during this difficult time, abortion advocates recently launched the abortion chatbot Charley,” The Register writes.
The Register claims that Charley offers poor advice surrounding abortion medications as Charley states that medications are available past FDA-approved gestational periods.
Perhaps сhatbots aren’t the way to source your abortion information, especially with digital privacy concerns?
You’re never alone online
In the age of digital surveillance, we are all being tracked in one way or another. This goes for all those using period tracking apps or any app for that matter.
Now, digital surveillance is being used in a post-Roe vs. Wade world to prosecute those seeking abortion.
The case of Latice Fisher from 2017, before the overturn, demonstrates how digital surveillance is being used to charge those suspected of having an abortion.
Fisher was indicted on the charge of second-degree murder after the stillbirth of her infant child, according to the Human Rights Crisis: Abortion in the United States After Dobbs report.
In the 2017 case, “mifepristone” and “misoprostol” were found in her phone’s search history, the report claims.
These are abortion pills commonly used to terminate a pregnancy.
Although it’s unknown as to whether she took the pill, law enforcement has used this data to charge the woman for murder, the Washington Post reported.
This occurred before the overturn of Roe vs. Wade, demonstrating the use of digital surveillance to track abortions well before restrictions and bans were in place.
Therefore, we can conclude that the use of digital surveillance is likely to increase with the overturn of Roe vs. Wade, the report claims.
Your personal information may be used to identify whether you have carried out an abortion:
- Location data: if an individual has visited an abortion clinic or other health facility
- Search histories: searches for medications for abortion, clinics, and general information on abortion
- Menstrual cycle tracking apps and communications data regarding pregnancy and abortion
Chatbots like Charley might seem like a good idea if you want quick and easy information surrounding abortion.
But with the rise of digital surveillance, an individual seeking abortion in US states that have banned termination could face legal issues.
More from Cybernews:
Subscribe to our newsletter