WhatsTheBigData

A 49-year-old Brazilian man sold all his belongings and travelled nearly 1,900 kilometres to meet his dream woman. But when he arrived, he discovered that the woman was an AI-generated character, not a real person.

Edimar Martins, a bricklayer from Maranhão in northeastern Brazil, travelled to Belo Horizonte, the capital of Minas Gerais, after becoming emotionally attached to a social media profile featuring a woman named Jessica Oliveira, also known as Jessica Fazendeira.

Martins told local broadcaster TV Mirante that he first came across the account in 2020. The profile reportedly had more than 300,000 followers and featured videos of a woman living on a farm. In the videos, she appeared to offer men a place to stay and financial support if they agreed to live with her.

Martins had been dealing with loneliness following his mother’s death and believed the woman could help him begin a new chapter in his life. One of the videos reportedly showed the character promising financial assistance and offering him a home on her farm. Believing he had found someone he could build a future with, Martins decided to travel to meet her in person.

ALSO READ: Why Are Men Using AI Girlfriends?

Brazilian Man Travels 1,900 Km to Meet AI Girlfriend, Discovers She Isn’t Real

According to the New York Post, Martins sold his belongings on October 9 to pay for his trip to Belo Horizonte. He travelled by bus for more than a day and reached the city on October 11.

He carried photographs of the woman and asked people if they knew who she was. However, no one could help him find her.

Martins later turned to a local radio station for help. It was there that he discovered the truth. The woman he wanted to meet was an AI-generated character, and no real person was behind the social media profile.

The revelation left Martins heartbroken. He had sold his belongings, spent his money and travelled nearly 1,900 kilometres, hoping to find a partner.

“It only brings sadness, discouragement, and depression. Many people wanted to find a relationship, so I came,” Martins said.

Creator Confirms Woman Was an AI-Generated Character

The person behind the social media account later admitted that the woman in the videos was created using artificial intelligence (AI). The creator said they had never spoken to Martins through the account. They also explained that the character was made to attract attention and generate business opportunities.

“I came here to state the obvious: I am made of artificial intelligence. I have never spoken to anyone with this profile,” the creator said.

The creator also denied inviting Martins to Belo Horizonte or planning to meet him, saying the situation had come as a surprise.

The incident shows how AI-generated characters can appear real on social media. Realistic videos and personal stories can make it difficult for viewers to tell whether they are looking at a real person or a computer-generated character.

For people looking for love online, this can create confusion, especially when an account does not clearly state that its photos or videos have been created using AI.

ALSO READ: AI Relationship Statistics: How AI Is Changing Dating

Brazilian Man Seeks Help After Discovering His AI Girlfriend Was Fake

After learning that the woman he wanted to meet did not exist, Martins was left stranded in Belo Horizonte with little money to return home. He approached local authorities for help. Social workers reportedly gave him food, access to shower facilities and a bus ticket so he could leave the city.

Despite the disappointment, Martins has not given up on finding a partner. He is now staying with a family member and hopes to return to Belo Horizonte to find someone with whom he can share his life.

“I didn’t find her, but I’ll find another one, and it will be in the same place I went, and I’ll go again,” Martins said.

Martins’ experience ended very differently from what he had expected. He had travelled a long distance hoping to find love, only to discover that the woman he had become attached to was an AI-generated character.

Martins is the third man in less than six months to travel to Belo Horizonte hoping to meet a woman who turned out to be AI-generated. This shows the risks of AI-generated people on social media. Realistic images and videos can make fictional characters look like real people, leading followers to believe their stories are genuine.

AI-generated deepfake nudes are becoming a growing concern in schools. Students can use AI platforms to turn ordinary photos of their classmates into fake nude images. They can then share these images through messaging apps and school groups. For the students targeted, this can lead to embarrassment, fear and a serious invasion of privacy.

A joint investigation by WIRED and Indicator documented 88 cases involving AI nudification tools across 28 countries, affecting 640 students. These incidents highlight the need for schools to strengthen their efforts to protect students from AI-generated sexual abuse.

In many cases, schools face difficulties investigating incidents, helping affected students and deciding how to deal with those responsible. The situation becomes more difficult as AI tools make it easier to create fake nude images without advanced technical skills.

This makes deepfake abuse more than a problem of how students use AI. It also raises questions about who is responsible for protecting children. Schools, technology companies and governments must also decide how to respond when these tools are used to harm students.

AI Deepfake Nudes of Students Are Not a Harmless Prank

AI Deepfake nude images of students should not be dismissed as jokes or harmless teenage behavior. When someone uses AI to create fake nude images of a classmate without their consent, the consequences can be serious.

Although the images are fake, the harm to the person targeted is real. Students may feel ashamed, anxious or exposed when they learn that someone has created or shared sexual images of them. They also lose control over how their faces and bodies are shown to others.

The impact can last long after the images first appear online. Victims may worry about who has seen the images, whether others have saved them or if they will be shared again in the future. Some students may find it difficult to face their classmates or feel safe at school.

The person creating these images may think it is a joke or a way to impress friends. But the student targeted has little control over who sees the images or how far they spread. This difference makes the situation even more distressing for victims.

Schools must take these incidents seriously instead of treating them as ordinary arguments between students. Creating or sharing sexual images of classmates without their consent should prompt a proper investigation and support for the students affected, even when AI has been used to make the images.

An image does not have to be real to cause real harm. Using AI to create fake nude images of someone without their consent can violate their privacy and dignity. Calling it a prank does not make the behavior acceptable.

AI Deepfake Nudes Have Affected at Least 640 Students Across 28 Countries

A joint investigation by WIRED and Indicator found 88 publicly reported cases of AI-generated sexual abuse involving school communities across 28 countries. At least 640 students were affected. The cases date back to 2023 and include incidents in which students allegedly used AI nudification tools to create fake nude images of their classmates.

North America had nearly 30 reported cases, while Europe recorded more than 20. South America had more than 10 cases, and Australia and East Asia together accounted for around a dozen more.

However, the actual number of cases could be much higher. The investigation used publicly available reports, so it may have missed cases dealt with privately by schools or police. Language barriers and limited media coverage in some countries may also mean that many cases were never reported publicly.

AI Nudification Tools Are Making Deepfake Abuse Harder to Stop

AI tools have made it much easier to create fake nude images. People who once needed image-editing skills can now use apps, websites and bots to produce sexual images from ordinary photographs.

This creates a serious problem for schools. Students do not need to build their own AI tools or understand how the technology works. In some cases, they only need a photo of a classmate and access to a service that can turn it into a fake nude image.

The problem can grow once an image is created. Students may share it in group chats or post it on social media. Others can save copies and spread them further, making it difficult to control where the images end up.

Telling students to be more careful about the photos they post online will not solve the problem. It shifts the responsibility onto potential victims instead of addressing the people who create and share these images without permission.

Girls should not have to stop posting photos, avoid school events or leave social media because someone might misuse their pictures. They should be able to share ordinary photographs without fearing that someone will turn them into fake nude images.

The responsibility lies with those who create and spread these images, as well as the companies that provide these tools. Technology companies need stronger safeguards to prevent misuse, while schools must take reports seriously and help protect affected students.

ALSO READ: AI Has Made Fake Nude Abuse Easier to Scale, New Berkeley Report Warns

Schools Need a Clear Plan to Handle AI Deepfake Abuse

Schools respond differently to AI deepfake abuse cases. Some incidents lead to police investigations or disciplinary action. In other cases, parents say school officials have not taken their complaints seriously.

Such differences can leave students wondering whether reporting an incident will make any difference. They may also give other students the impression that creating or sharing fake nude images is not a serious offence.

Schools need clear rules for handling these cases. Staff should know what to do when a student reports abuse, how to keep evidence safe and when to involve parents, child protection services or the police.

Protecting the student targeted by the abuse must also be a priority. They should not have to explain what happened repeatedly to different staff members or face unwanted questions from classmates. Schools must also ensure that victims are not blamed, punished or made to miss classes because of the incident.

Teachers and school administrators need proper training to recognise AI-generated sexual abuse and understand how these images can spread. They should know how to respond to reports, protect students’ privacy and provide support without making the situation more difficult for the victim.

Schools should also teach students about the harm caused by creating or sharing sexual images of others without their permission. Students need to understand that using AI to create these images does not make the behaviour acceptable and that it may have legal consequences.

However, simply warning students about the risks will not be enough. Schools need reliable ways to report abuse, staff who know how to handle complaints and a clear plan to protect students when an incident occurs.

Social Media Platforms and AI Companies Must Take Responsibility

Schools often have to deal with the damage caused by abuse that starts or spreads online. This means AI companies and social media platforms also have a role to play in protecting students.

AI companies should take steps to stop their tools from being used to create sexual images of real people without their consent, particularly children. They also need to address attempts to get around these safeguards and make it easy for users to report harmful content.

Social media platforms have a separate responsibility to act when someone reports a fake nude image. Removing the original post may not be enough if other users have already saved the image and shared it elsewhere. Platforms need ways to limit its further spread and respond to reports without unnecessary delays.

Some laws are beginning to address this problem. In the United States, the TAKE IT DOWN Act requires covered platforms to offer a way to report nonconsensual intimate images, including AI-generated images. After receiving a valid request, platforms must remove the reported content and known identical copies within 48 hours.

The law is an important step towards protecting victims, but removing images after they appear online cannot prevent every case of abuse. Schools, parents and technology companies still need to work on prevention and respond when incidents occur.

AI companies should not wait until someone becomes a victim before reviewing the risks of their products. They need to build safeguards into their services from the start and take action when those protections fail. Preventing the misuse of AI should be a basic responsibility, not something companies address only after harm has occurred.

ALSO READ: Meta Says Its AI Caught 97% of Child Exploitation Content Before User Reports

Protecting Students Takes More Than Banning AI Tools

Banning AI tools in schools may seem like a simple way to prevent deepfake abuse, but it will not solve the problem on its own. Students can still access these services outside school, and fake nude images can spread through personal phones, social media and private messaging groups.

A better approach is to focus on preventing abuse, holding those responsible accountable and supporting the students affected.

Schools need clear rules against creating and sharing fake sexual images of classmates. They should also provide safe, confidential ways for students to report incidents. Students must understand that using AI to create these images without consent can hurt others and may have serious consequences.

Parents also have a role to play. Talking to children about consent, responsible online behaviour and the harm caused by sharing intimate images without permission can help them understand why this behaviour is unacceptable.

Governments need to ensure that laws protect people whose images are misused and give victims a way to seek help. AI companies and social media platforms must strengthen their safety measures and respond quickly when abuse is reported. Schools, in turn, should make sure that students who report these incidents receive support instead of being blamed for what happened.

There is no single solution that can stop every case of AI deepfake abuse. However, schools, parents, governments and technology companies can work together to prevent harm, hold those responsible accountable and help victims get the support they need.

Protecting Victims Must Come First in AI Deepfake Cases

AI-generated deepfake nudes have exposed how difficult it can be for schools and other institutions to keep up with new technology and deal with the harm it can cause.

The technology may be new, but using images to shame, threaten or hurt someone is not. AI has made it easier to create fake sexual images, while many schools are still figuring out how to handle these cases.

The reported incidents across dozens of countries show that this is not just a problem affecting a few schools. The available data only includes those cases that have become public, so many incidents may go unreported or remain unknown.

Students should not have to stop sharing photos or avoid social media to protect themselves from this abuse. They should also not be expected to fight for justice on their own while dealing with the emotional impact of what happened.

Schools need a clear plan for handling these incidents before another student becomes a victim. AI companies must take steps to prevent the misuse of their tools, and social media platforms need to act quickly when harmful images are reported.

The main goal should not be how quickly a school deals with public attention after an incident. It is whether the school protects the student, holds those responsible accountable and takes steps to stop the same thing from happening again.

AI-generated nude images involving children are on the rise. It involves using ordinary photographs to create fake nude or sexual images of children without their consent. These images can be shared on social media or among classmates. They can also be used to bully, embarrass or threaten the child. The person does not need an actual intimate photograph to cause harm.

The problem is becoming harder to ignore. In the UK, more children are reporting fake or digitally altered sexual images of themselves. The rising number of reports has raised concerns among child protection groups, schools and families. They are worried about how easily these images can be created and shared.

According to The Guardian, the Report Remove service received 420 reports from children in the first six months of 2026. These reports involved suspected fake or manipulated explicit images of the children themselves. The numbers were already higher than the 397 reports recorded during the whole of 2025.

AI deepfake abuse is a child safety issue, not just a technology problem. Children should not have to limit the photos they share online to protect themselves from abuse. Schools and families must help children who face this problem. Technology companies and child protection services also have a role to play. They must help victims, remove abusive images and prevent further harm.

ALSO READ: AI Porn Statistics (2025-2030): Key Data, Trends & Facts

AI Nudification Tools Are Making Sexual Image Abuse Easier

AI nudification tools have made it easier to create fake nude images of people without their consent. In the past, creating realistic fake images often required special software and image-editing skills. Today, some websites, apps and bots can turn ordinary photos into fake sexual images with little effort.

These tools are especially concerning because they can turn a photo of a fully clothed person into a fake nude image. The person in the original photograph does not need to have taken or shared an intimate picture for this abuse to happen.

For children, this creates another risk online. A school photograph, a selfie posted on social media or another publicly available picture can be used to create a fake sexual image. Although the image is not real, the embarrassment, bullying and emotional distress it causes can have serious consequences.

The abuse can continue even after the image has been created. Someone may share it in a school group chat, post it on social media or send it directly to other students. Others can then download and share the image again, making it difficult to stop its spread. Removing the original post may not prevent further sharing.

In August 2026, the Internet Watch Foundation (IWF) reported an increase in children contacting the Report Remove service. They were seeking help with nude or sexual images they believed had been faked or digitally altered.

Not every case of sexual image abuse involves AI. Some images are created using traditional editing software or other manipulation methods. However, AI nudification tools are becoming more widely available. This makes it easier to create fake sexual images of children and increases the risks they face online.

Children Should Not Have to Sacrifice Their Privacy to Stay Safe Online

One of the biggest concerns about AI-generated sexual abuse is that children can become victims through no fault of their own. Someone can use a photo shared with family or posted on social media to create a fake nude image. This can happen without the child’s knowledge or consent.

The Internet Watch Foundation (IWF) has warned that ordinary photographs can be manipulated to create sexual images. This puts parents in a difficult position. Many families share pictures of their children to celebrate birthdays, school events and other special moments. However, they may not realise that publicly available photographs can be used to create fake sexual content.

In July 2026, The Guardian reported on warnings from the UK’s National Crime Agency and the IWF about the risks of sharing children’s photographs publicly. The guidance highlighted concerns that AI tools could allow offenders to create sexual abuse material without having direct contact with a child. Parents can reduce the risks by taking a few precautions:

• Review privacy settings on social media accounts.
• Limit who can view their children’s photographs.
• Talk to children about online safety and the risks of sharing photos publicly.
• Remember that these steps can reduce risks but cannot completely prevent image misuse.

Protecting children from AI deepfake abuse should not mean asking them to stop sharing photographs or blaming parents when abuse occurs. Children deserve to take part in online life without having to fear that an ordinary photograph could be turned into a fake sexual image. Parents should also be able to share family memories without carrying the full burden of preventing this type of abuse.

The responsibility must extend beyond families. Companies that develop AI tools must take steps to prevent their misuse, social media platforms must respond quickly to reports of abusive images, and authorities must work to protect children and hold offenders accountable. Keeping children safe online requires action from everyone involved, not just the people who may become victims.

Online Platforms Must Make It Easier to Report Fake Nude Images

When a fake nude image of a child appears online, getting it removed can be difficult. Children may not know where to report the image, what information to provide or whether asking for help will make the situation worse. Some may also fear being blamed for an image they never created or shared.

The UK’s Report Remove service gives young people a way to seek help. Run by the Internet Watch Foundation (IWF) in partnership with the NSPCC’s Childline service, it allows children under 18 to report nude or sexual images of themselves. Specialists can assess the reported material and work to have eligible images removed from the internet.

The service gives children a way to report sexual image abuse without having to handle the situation alone. The growing number of reports also highlights the need for better support for young people whose photographs have been turned into fake sexual images or manipulated without their consent.

However, reporting services cannot stop every case of online sexual abuse. An image may have already been copied, uploaded to multiple websites or shared through private messaging groups. Some children may also avoid reporting abuse because they fear embarrassment, punishment or further bullying from classmates.

Social media platforms and websites need simple, accessible ways for victims to report fake sexual images. They should review reports quickly and remove abusive content where necessary. They should also prevent the same images from being uploaded again where possible. In addition, they must protect the personal information of affected children.

AI Companies Must Take More Responsibility for Preventing Deepfake Abuse

AI companies must take steps to stop their tools from being used to create fake sexual images of children. Safety checks should be built into AI image generators before they are released.

Companies should test their tools, block requests to create sexual images of minors and fix weaknesses that allow users to bypass safety rules. They should also take action against services that create fake sexual images of people without their consent.

These safety measures must cover third-party apps, bots and modified versions of AI tools. Blocking harmful requests on one platform is not enough if similar services remain available elsewhere.

Social media platforms must also act quickly when someone reports a fake sexual image. They should remove abusive content where appropriate and protect the privacy of affected children. They should take reports seriously even when it is difficult to confirm how an image was created.

No single safety measure can prevent all cases of AI deepfake abuse. Companies need strong safeguards, simple reporting options and quick responses when abuse occurs.

In November 2025, the UK government announced legislation aimed at preventing AI models from being used to create child sexual abuse material. The proposed measures would allow authorised organisations to test AI systems for safety risks.

However, laws and safety promises are not enough on their own. AI companies must show that their safeguards work and act when their tools are used to harm children.

ALSO READ: UK Bans Ads for AI Nudify Apps That Sexualize Real Women

Laws Must Keep Pace With AI-Generated Sexual Abuse

Laws must address the creation and sharing of AI-generated sexual images of children. Even when an image is fake, it can cause serious harm to the child whose photograph has been used.

Governments need clear laws to deal with this abuse. Police also need the skills and resources to investigate cases, identify offenders and work with online platforms to remove harmful images.

International cooperation is important because these images can be created in one country and shared across the world. Differences in laws can make it harder to stop their spread and hold offenders accountable.

Children who experience AI deepfake abuse should never be blamed for what happened. They need safe ways to report abuse and access support. Stronger laws must help prevent abuse, protect children and hold offenders accountable.

Children Facing AI Deepfake Abuse Need Support, Not Blame

The harm caused by AI deepfake abuse does not end when an image is removed. Children may worry that their classmates have seen it, that it will appear online again or that others will believe it is real. Some may feel ashamed or blame themselves, even though they did nothing wrong.

Parents, teachers and other trusted adults should take these concerns seriously. They must avoid blaming children for sharing a photograph or using social media. The responsibility lies with the person who created or shared the abusive image.

When a child reports AI-generated sexual images of themselves, adults should help them report the abuse and seek support from child protection services or the police when needed. They should also keep relevant information that may help an investigation without sharing or circulating the image further.

Schools must protect affected students from bullying and provide access to emotional support. Students who create or share abusive images may need disciplinary action, but schools must also focus on the safety and well-being of the child targeted.

Every child should feel safe asking for help. Children facing AI deepfake abuse need protection, understanding and support, not blame for someone else’s actions.

AI Innovation Must Not Come at the Cost of Children’s Safety

The rise in reports to the UK’s Report Remove service shows that AI deepfake abuse is a growing concern. AI tools can make it easier to create fake sexual images, leaving children to deal with the harm.

Report Remove received 420 reports of suspected fake or manipulated sexual images in the first six months of 2026, compared with 397 during all of 2025. These figures do not show the total number of victims, and not every case necessarily involved AI. However, they show that more children are reporting such images.

Although the full scale of the problem is unclear, that should not stop action. AI companies must make their tools safer. Social media platforms should respond quickly to reports of abusive images. Schools need clear rules for handling these cases, and governments must ensure that laws keep pace with new technology.

Parents can teach children about online safety, but they cannot prevent every case of abuse. Children should not have to stop sharing photos online out of fear that someone might misuse them. AI companies must do more than create realistic images. They must also take steps to prevent their tools from being used to harm children.

Protecting children from AI deepfake abuse requires action from technology companies, online platforms, schools and governments. Children and their families should not have to carry this responsibility alone.

AstraZeneca has signed a deal worth up to $1.5 billion with biotechnology company Insitro to use artificial intelligence (AI) to discover new treatments for liver and heart diseases. Announced on October 15, 2026, the partnership will use Insitro’s “Virtual Human” AI platform to study diseases and identify potential targets for new drugs.

The research will focus on metabolic dysfunction-associated steatohepatitis (MASH), a serious form of fatty liver disease, and cardiomyopathies, which affect the heart muscle. The deal reflects growing interest in using AI to speed up drug discovery.

AstraZeneca will pay insitro $40 million upfront. Insitro could receive another $1.46 billion if the research meets specific development and sales targets. The deal also includes royalties on future drug sales, meaning the full $1.5 billion will depend on the programmes’ progress and commercial success.

What Is Insitro’s Virtual Human Platform?

Insitro’s Virtual Human platform uses artificial intelligence (AI), machine learning and lab-grown human cells to study how diseases develop. It combines biological data with computer-based analysis to help scientists understand the causes of diseases and identify possible targets for new drugs.

Finding the right drug target is an important step in drug discovery. It helps researchers identify which parts of the body or biological processes a new medicine should affect.

Under the AstraZeneca partnership, Insitro will use the platform to study MASH and cardiomyopathies. Researchers will investigate targets identified by AstraZeneca and work to turn promising findings into potential drug candidates.

The Virtual Human platform is not a complete digital copy of a person. It uses biological models, data and machine learning to study human diseases. Any potential findings will need further testing before they can support the development of new medicines.

ALSO READ: Mark Zuckerberg Bets $1.8 Billion on AI That Could Build Virtual Human Cells

AstraZeneca Turns to AI to Find New Treatments for Liver and Heart Diseases

Developing new medicines can take years, partly because scientists need to understand what causes a disease and how to treat it. AI can help researchers analyse biological data and identify potential drug targets earlier in the research process.

AstraZeneca will use insitro’s AI platform to study MASH and heart muscle diseases. The aim is to identify biological targets that could help researchers develop new treatments for patients with these conditions.

The two companies will share the work. Insitro will use its AI technology and laboratory research to study diseases and identify potential drug candidates. AstraZeneca will handle clinical development and the later stages of bringing successful medicines to market.

The partnership combines Insitro’s expertise in AI and biological research with AstraZeneca’s experience in developing medicines. This could help researchers identify new treatment options for liver and heart diseases.

ALSO READ: AI Is Changing How Scientists Search for Treatments for Brain Diseases

AstraZeneca and Insitro Target Fatty Liver Disease and Heart Conditions

AstraZeneca and Insitro will use AI to identify potential drug targets for two conditions: metabolic dysfunction-associated steatohepatitis (MASH), a serious form of fatty liver disease, and cardiomyopathies, which affect the heart muscle.

MASH causes fat buildup, inflammation and liver damage. In severe cases, it can lead to cirrhosis or liver failure. The companies will use AI and human cell models to study the disease and identify possible targets for new treatments.

The research will also examine cardiomyopathies, which can affect the heart’s ability to pump blood and may lead to heart failure. The aim is to understand the biological causes of these conditions and identify potential drug candidates.

The research is still in its early stages. Any potential drugs will need further research and testing before they can be made available to patients.

How AstraZeneca’s $1.5 Billion AI Deal Could Shape Drug Discovery

The $1.5 billion AstraZeneca-insitro deal highlights the growing use of AI in drug discovery. The partnership combines AI analysis with laboratory research to help scientists understand diseases and identify potential targets for new medicines.

The agreement also shows that pharmaceutical companies are investing in biotechnology firms to support early-stage drug research. However, Insitro will receive most of the potential payments only if the research programmes meet specific development and commercial milestones.

The success of the partnership will depend on whether the research identifies promising drug targets and leads to potential treatments. Any resulting drug candidates will still need further development and testing.

The deal reflects AstraZeneca’s investment in AI-based drug discovery, but it remains uncertain whether the research will produce new medicines for liver and heart diseases.

The Group of Seven (G7) has agreed to run simulated AI cyberattacks. It will test how governments, central banks and financial regulators respond to cyber threats. The exercises will focus on protecting banks and key financial systems from attacks involving advanced artificial intelligence (AI).

The move follows a warning from the International Monetary Fund (IMF). The IMF said an AI-based cyberattack on a critical financial service could weaken public trust in financial institutions.

Japanese Finance Minister Satsuki Katayama confirmed the plan after a meeting of G7 finance ministers and central bank governors. The meeting took place alongside the IMF’s annual meetings in Bangkok, Thailand. Reuters reported the agreement on October 16, 2026.

G7 to Test Responses to AI Cyberattacks

The planned exercises will simulate cyberattacks involving advanced artificial intelligence (AI). They will help financial authorities test their emergency response plans and assess their readiness for such threats.

Officials will review the results to identify gaps in existing security measures. This might also lead to updates to response guidelines for handling AI-related cyberattacks. G7 financial authorities will develop the exercises. Private financial institutions may also take part, depending on the scenarios.

The G7 consists of Canada, France, Germany, Italy, Japan, the United Kingdom and the United States. The group works together on economic and financial issues affecting its members and the global economy.

Why AI Cyberattacks Are a Growing Threat to Banks

Banks and other financial institutions face new cybersecurity risks as AI systems become more advanced. Attackers could use AI to find software weaknesses, automate cyberattacks and carry out attacks faster.

These threats could affect banks, payment services and other critical financial systems. An attack on one institution could also spread through connected networks and disrupt services in multiple countries.

In a September 2025 statement on AI and cybersecurity, the G7 Cyber Expert Group warned about the risks of AI misuse. It said AI could increase the frequency and impact of cyberattacks. It also highlighted risks arising from financial institutions’ reliance on shared technology providers.

AI can also help banks detect security weaknesses, identify suspicious activity and respond to cyber incidents. However, financial authorities must prepare for attackers to use the same technology to target banks and critical financial systems.

ALSO READ: AI Cyber Attacks Statistics [2025 and beyond]

IMF Warns AI Cyberattacks Could Threaten Financial Stability

The International Monetary Fund (IMF) has warned that AI-related cyberattacks could pose risks to financial stability. An attack on a critical financial service could disrupt operations and weaken public trust in banks and other financial institutions.

The threat extends beyond individual banks. Financial institutions depend on connected networks, technology providers and payment systems. A major cyberattack could affect multiple institutions and disrupt financial markets at the same time.

The G7’s planned cyberattack simulations will test how authorities coordinate their response to such incidents. The exercises could help identify gaps in communication, crisis management and recovery plans before a real attack takes place.

ALSO READ: ARTEX AI Agent Pulled From GitHub After Cyberattacks Target South Korean Banks

G7 Strengthens Cybersecurity Measures Against AI Threats

The G7 is taking further steps to prepare banks and financial systems for new cyber threats. Regulators and central banks are assessing how advanced artificial intelligence (AI) could affect existing security measures and the ability of financial institutions to recover from major cyberattacks.

In July 2026, the G7 Cyber Expert Group announced that it had completed a cross-border exercise simulating a large-scale cyberattack across G7 countries. The exercise involved finance ministries, central banks, banking supervisors and financial market authorities. It tested how these agencies respond to cyber incidents, share information and communicate during a crisis.

The group also adopted a long-term plan to hold these exercises more regularly and follow a consistent approach. The upcoming AI cyberattack simulations will build on these efforts. They will test how financial authorities respond to cyber threats involving advanced AI systems.

How G7 Plans to Test Banks Against AI Cyberattacks

The planned G7 exercises could help governments and financial institutions find weaknesses in their cybersecurity plans and improve coordination during a crisis.

The tests will also help authorities assess whether current security measures can handle attacks involving advanced AI systems. However, their effectiveness will depend on the attack scenarios used. It will also depend on how officials use the findings to improve security.

The G7 has not yet shared a detailed schedule or confirmed which cyberattack scenarios the exercises will cover.

As AI technology advances, testing how financial systems respond to cyberattacks will become increasingly important. These tests can help protect banks, payment networks and the global economy.

Bank of America has hired 1,000 outside contractors to fix software security flaws found by advanced artificial intelligence (AI) models, including Anthropic’s Mythos. The move shows how AI tools can identify vulnerabilities faster than companies can fix them.

Bank of America CEO Brian Moynihan revealed the additional hiring in October 2026. The new contractors joined around 3,000 existing employees at the bank. Their work involves finding and fixing security flaws in Bank of America’s systems and third-party software.

The bank began hiring the contractors in the second quarter of 2026. They were still working with the company in October as Bank of America continued to address vulnerabilities identified by AI models.

Why Bank of America Hired 1,000 Contractors

Bank of America hired additional contractors after advanced AI models found software vulnerabilities that needed further checks and repairs. Anthropic’s Mythos can identify software flaws and assess how attackers could exploit them.

These AI capabilities can help security teams find weaknesses that traditional testing methods may miss. However, finding a vulnerability is only the first step. Teams must examine each issue, decide which ones need urgent attention and fix them without disrupting banking services.

Bank of America needed more technical staff to handle the growing workload. The 1,000 contractors are helping existing employees fix vulnerabilities in the bank’s software and applications developed by third-party vendors.

ALSO READ: AI Cyber Attacks Statistics [2025 and beyond]

Anthropic’s Mythos Raises Cybersecurity Concerns

Anthropic introduced Mythos in April 2026 with advanced capabilities to identify software vulnerabilities. The company limited access to the model because its ability to find and exploit security flaws could also create risks if attackers used it for cyberattacks.

Mythos can analyse software code, detect weaknesses and identify ways to exploit them. This creates a challenge for businesses that must fix security flaws before cybercriminals can take advantage of them.

To address these risks, Anthropic launched Project Glasswing. The initiative gives selected organisations access to its cybersecurity tools to help them find and fix weaknesses in critical software.

Banks face particular pressure to address these security risks. Financial institutions rely on complex IT systems and handle sensitive customer data. Any unpatched vulnerability could expose these systems to potential attacks.

Reuters reported in May 2026 that US banks were working to fix IT vulnerabilities identified by Mythos. This has raised questions about whether financial institutions could repair these weaknesses quickly enough. Advanced AI tools have made it easier to discover software flaws.

How AI-Detected Security Flaws Increase Cybersecurity Costs

Identifying software vulnerabilities is only the first step in cybersecurity. Once a flaw is identified, security teams must assess its severity, understand the risks and find a way to fix it.

The teams must then test the fixes to make sure they do not create new security problems or disrupt existing services. Large banks also need to coordinate repairs across their own systems and software supplied by external vendors.

This work can take significant time and resources, particularly when AI tools uncover large numbers of vulnerabilities across extensive amounts of code.

Bank of America’s decision to hire 1,000 contractors shows how large-scale AI security testing can increase the demand for technical staff. The additional workers are helping the bank investigate and fix vulnerabilities while its existing teams continue their work.

During discussions about the bank’s third-quarter results, CEO Brian Moynihan also addressed the costs involved in fixing these issues. However, the reported spending covered other areas, including anti-money-laundering matters. Therefore, it should not be treated as the confirmed cost of fixing vulnerabilities identified by AI models alone.

Bank of America Highlights the Challenges of AI Cybersecurity

Bank of America’s decision highlights a challenge facing businesses that use advanced AI cybersecurity tools. These tools can find software vulnerabilities quickly. But companies also need enough skilled workers and resources to fix the problems.

For banks, delays in fixing serious security flaws can increase the risk of cyberattacks, data leaks and service disruptions. Financial institutions may need to invest in automated security testing. They may also need to strengthen the teams responsible for fixing the vulnerabilities these tools uncover.

AI tools may eventually help security teams decide which vulnerabilities to fix first and automate some repairs. However, experts still need to test these fixes carefully. Mistakes in banking systems could affect customers and disrupt essential financial services.

ALSO READ: OpenAI Agent Accessed Non-Public Files on Australian Government Medicare Portal

Bank of America’s experience shows why companies must prepare for the workload created by advanced AI security tools. These tools can find software weaknesses faster. However, existing teams may struggle to fix all the problems they uncover.

As AI cybersecurity tools improve, businesses will need to strengthen their ability to identify and assess software vulnerabilities. They must also ensure they have enough staff to fix these issues.

CoreWeave’s falling shares and bond prices have raised questions about the money being used for AI. The company, backed by Nvidia, has around $50 billion in debt. Investors are concerned that its future earnings may not be enough to pay its bills and fund further growth.

On October 15, 2026, CoreWeave stock shares fell again, even as CEO Mike Intrator expressed confidence in the company’s customer contracts. Reuters reported that the stock had dropped more than 15% on Wednesday, followed by a further 6% decline in premarket trading on Thursday.

The company’s troubles also reflect a wider concern across the AI industry. Tech companies are spending billions on data centers, powerful chips and computing equipment. Some are borrowing heavily to pay for these investments.

Investors are now questioning whether the money earned from AI services will cover these costs. As borrowing becomes more expensive, companies face greater pressure to manage debt and show their investments can generate profits.

ALSO READ: AI Debt Issuance Falls to $23 Billion as Investors Reconsider the AI Boom

CoreWeave Faces Growing Concerns Over Its $50 Billion Debt

CoreWeave provides computing power to companies that develop and run AI models. To meet rising demand, it spends heavily on Nvidia chips, data centers, and other equipment. The company relies on borrowed money to fund much of this expansion, adding to its debt burden.

Investors are now questioning whether CoreWeave can manage its debt as spending continues to rise. Yields on its bonds due in 2031 have climbed above 16%, suggesting that investors see greater risks in lending to the company.

According to Reuters, the cost of insuring CoreWeave’s debt against default has also risen sharply. The latest figures indicated a nearly 63% chance of default within five years, compared with about 30% a year earlier.

Strong AI Demand Has Yet to Ease Concerns About CoreWeave’s Profits

CoreWeave’s financial troubles show that strong demand for AI services does not always lead to higher profits. Even when customers need more computing power, companies must manage the high costs of building and maintaining the infrastructure that supports them.

Analysts cited by Reuters expect CoreWeave to bring in nearly $13 billion in revenue in 2026. However, the company is also projected to post a loss of about $2.13 billion. Its spending plans have added to investor concerns. In August, CoreWeave raised its forecast for annual capital spending by $4 billion, putting the expected total at up to $39 billion.

The company needs this money to expand its data centers, purchase chips and meet the computing needs of its customers. These investments are necessary to support its business, but they also increase the amount of money it needs to raise.

This leaves CoreWeave facing a difficult balance. It must keep investing to meet growing demand, while managing its debt and controlling its costs.

Nvidia’s Support Gives CoreWeave Some Relief

CoreWeave has major customers, including Microsoft and OpenAI, and strong ties with Nvidia. According to Reuters, Nvidia has agreed to buy any computing capacity CoreWeave cannot sell to other customers through April 13, 2032.

CEO Mike Intrator told employees in a memo that CoreWeave’s customer contracts were worth more than its debt. Still, this does not mean CoreWeave is free from financial trouble. The company is also seeking an investment-grade credit rating, co-founder Brannin McBee told the Financial Times. A higher credit rating could make it easier for CoreWeave to borrow money.

Gartner Warns of Risks From Heavy AI Infrastructure Spending

CoreWeave’s financial troubles have added to concerns about whether revenue from AI services can justify the huge cost of building and running data centers. On October 15, Gartner warned that around 80% of specialised cloud providers, known as neoclouds, could shut down by 2030. The firm also said more than $3.5 trillion in AI infrastructure investments could lose value if companies fail to generate sufficient returns. 

Gartner expects AI infrastructure spending to reach nearly $1 trillion in 2026, followed by another $6 trillion between 2027 and 2030. Analyst Will Sommer warned that revenue from AI services may not be enough to make these investments profitable. If demand falls short of expectations, companies could struggle with expensive infrastructure and rising debt payments.

ALSO READ: AI Infrastructure Spending Statistics

AI Debt Concerns Spread Across Technology Stocks

CoreWeave stock price has raised concerns across the AI industry, affecting other companies investing heavily in AI infrastructure. Reuters reported that shares of rival cloud providers Nebius, Applied Digital and IREN fell between 3.2% and 5% on Thursday.

The wider tech sector has also suffered losses. The Philadelphia Semiconductor Index dropped 12% over six sessions, wiping out around $1.6 trillion in market value. The Nasdaq also fell more than 3% during the same period.

Investors are increasingly questioning the cost of the AI boom. Major US technology companies are expected to spend more than $700 billion on infrastructure in 2026, according to Reuters.

Companies such as Alphabet and Meta are also facing questions about the returns on their AI investments. Even businesses with strong finances could face pressure if spending continues to rise faster than revenue.