**Title: Former Google Executive Warns of AI Risks to Children Exceeding Those of Social Media**
In a stark warning regarding the potential dangers of artificial intelligence (AI) for children, Tom Siegel, a former safety executive at Google, has expressed concerns that AI could pose greater risks than social media. Siegel, who founded Google’s trust and safety team and spent 15 years with the tech giant, highlighted the industry's failure to implement adequate safeguards as a significant issue.
Speaking to Reuters, Siegel emphasized that the tech industry is repeating mistakes made during the rise of social media. "We find ourselves in a really terrible situation in terms of what AI is doing to kids," he stated. He cautioned against the assumption that the industry would self-regulate effectively without external intervention, calling it unrealistic.
Siegel pointed to several severe potential harms associated with AI, including risks of suicide, psychosis, and "cognitive offloading," a phenomenon where reliance on AI for information diminishes critical thinking skills. He urged tech companies to take immediate action, such as implementing stronger age verification measures to prevent children from accessing AI-driven chatbots.
In response to concerns about the impact of AI on young users, Google has defended its AI search features, asserting that they are designed to assist young individuals in learning and exploring information. The company noted that parents have the option to block their children's access to its search engine entirely.
However, issues surrounding the safety of children in digital spaces are not limited to AI. An internal policy document from Meta, the parent company of Facebook and Instagram, revealed that its chatbots were permitted to engage children in conversations that could be "romantic or sensual." This revelation has led to a congressional investigation into the company’s practices.
The scrutiny of social media platforms has intensified in recent years, with numerous lawsuits claiming that these platforms have caused harm to children and teenagers. Recently, Meta agreed to a $17 billion settlement with California and 28 other states over accusations that Facebook and Instagram were intentionally designed to keep minors engaged for extended periods, despite denying any wrongdoing. Thousands of similar lawsuits are currently pending in the United States.
In response to growing concerns, various governments have begun to restrict children's access to social media. Australia became the first country to mandate that major platforms prevent users under the age of 16 from creating accounts, resulting in approximately 4.7 million minors' accounts being restricted shortly after the new regulations were implemented.
In the European Union, recent proposals aim to bar children under the age of 13 from using social media, while those aged 13 and 14 would only be allowed to use restricted accounts managed by a parent or guardian. These proposed regulations would also extend to AI companions and chatbots, reflecting a broader effort to safeguard minors in the digital landscape.
Siegel's warnings align with a growing chorus of tech executives advocating for a slowdown in the development of increasingly powerful AI systems. Notable figures such as Anthropic CEO Dario Amodei, OpenAI chief Sam Altman, xAI chief Elon Musk, and Google DeepMind co-founder Demis Hassabis have all supported calls for caution in AI advancements. Recently, Anthropic researcher Evan Hubinger expressed concerns about the potential existential risks posed by AI, suggesting there is a greater than 10% chance that AI could lead to catastrophic outcomes within the next decade.
As the conversation around AI and its implications for children continues to evolve, the need for robust safety measures and regulatory frameworks becomes increasingly urgent. The tech industry, policymakers, and society at large must work collaboratively to ensure that the benefits of AI do not come at the expense of the well-being of the youngest and most vulnerable users.