The integration of artificial intelligence (AI) into computational protein design has significantly enhanced the accuracy and power of this field, promising transformative advances in biotechnology, particularly in sustainability and medicine. However, this technology also poses risks, including the potential for misuse and the creation of dangerous biological agents. To ensure safety, all synthetic gene sequences and synthesis data should be stored in secure repositories, accessible only in emergencies. AI-accelerated protein design can address global challenges such as pandemics, neurodegenerative diseases, and ecosystem degradation. Machine learning enables the rapid creation of novel biomolecules with diverse structures and functions. Advances in DNA synthesis have made it easier to encode these proteins into synthetic genes. Last year, the first drug developed through computational protein design, the COVID-19 vaccine SKYCovione, was approved internationally. However, regulating AI is a complex process, and overly restrictive regulations could hinder progress. A conference in Seattle highlighted the need for biosecurity measures, including screening and logging all synthesized genetic sequences. This would prevent the creation of harmful biomolecules. Since 2004, the International Gene Synthesis Consortium (IGSC) has regulated DNA synthesis, screening sequences for homology to pathogen components. Future checks could link sequence screening with synthesis itself, using cryptographic methods to ensure security. Logging sequences is essential, even if they have no similarity to natural proteins, to trace new biological threats. A "selective revelation" policy would allow queries only under exceptional circumstances. Standardizing screening and logging practices internationally is crucial. Security should not hinder information sharing, and biosecurity should be framed as a means to maximize progress in addressing societal concerns. David Baker and George Church are leading experts in this field, emphasizing the need for collaboration among all relevant communities to ensure the safe and effective use of AI in protein design.The integration of artificial intelligence (AI) into computational protein design has significantly enhanced the accuracy and power of this field, promising transformative advances in biotechnology, particularly in sustainability and medicine. However, this technology also poses risks, including the potential for misuse and the creation of dangerous biological agents. To ensure safety, all synthetic gene sequences and synthesis data should be stored in secure repositories, accessible only in emergencies. AI-accelerated protein design can address global challenges such as pandemics, neurodegenerative diseases, and ecosystem degradation. Machine learning enables the rapid creation of novel biomolecules with diverse structures and functions. Advances in DNA synthesis have made it easier to encode these proteins into synthetic genes. Last year, the first drug developed through computational protein design, the COVID-19 vaccine SKYCovione, was approved internationally. However, regulating AI is a complex process, and overly restrictive regulations could hinder progress. A conference in Seattle highlighted the need for biosecurity measures, including screening and logging all synthesized genetic sequences. This would prevent the creation of harmful biomolecules. Since 2004, the International Gene Synthesis Consortium (IGSC) has regulated DNA synthesis, screening sequences for homology to pathogen components. Future checks could link sequence screening with synthesis itself, using cryptographic methods to ensure security. Logging sequences is essential, even if they have no similarity to natural proteins, to trace new biological threats. A "selective revelation" policy would allow queries only under exceptional circumstances. Standardizing screening and logging practices internationally is crucial. Security should not hinder information sharing, and biosecurity should be framed as a means to maximize progress in addressing societal concerns. David Baker and George Church are leading experts in this field, emphasizing the need for collaboration among all relevant communities to ensure the safe and effective use of AI in protein design.