#security #ai #llm Created at 230923 # [Anonymous feedback](https://www.admonymous.co/louis030195) # [[Epistemic status]] #shower-thought Last modified date: 230923 Commit: 0 # Related # production software security with ai ## aerospace AI and software safety in space exploration, such as rockets and satellites, are crucial for ensuring efficient and secure operations. Some key aspects of AI safety in the space industry include: 1. **Autonomous Spacecraft Operations**: AI is used to automate complex tasks, enhance efficiency, and improve overall spacecraft design and construction processes[14]. 2. **Rocket Design and Navigation**: AI plays a crucial role in optimizing rocket design, trajectory, and speed, as well as automating certain aspects of rocket landing procedures[10]. 3. **Satellite Technology**: AI-controlled satellites have the potential to enhance satellite autonomy and safety, improving various aspects of satellite operations, such as relative positioning, communication, and end-of-life management[5]. 4. **Software Safety Standards**: NASA has developed software safety standards, such as NASA-STD-8739.8B, to provide a systematic approach to software assurance, software safety, and Independent Verification and Validation (IV&V) for software used in space missions[7][11]. 5. **Machine Learning Security Operations (MLSecOps)**: The Space ISAC AI/ML community has published a whitepaper on MLSecOps, which provides insight and considerations for using trustworthy AI and machine learning security operations to protect national space assets[8]. 6. **Collaboration and Information Sharing**: Companies and organizations in the space industry are collaborating and sharing information on managing AI risks, best practices for safety, and technical collaboration to ensure the safe and secure implementation of AI technologies[12]. By integrating AI technologies into space exploration, the industry can enhance efficiency, safety, and overall performance, pushing the boundaries of space exploration and scientific discovery[14]. Citations: [1] https://spacenews.com/op-ed-the-dramatic-influence-of-generative-ai-on-the-space-industry/ [2] https://www.forbes.com/sites/bernardmarr/2023/04/10/artificial-intelligence-in-space-the-amazing-ways-machine-learning-is-helping-to-unravel-the-mysteries-of-the-universe/?sh=24acfc847b60 [3] https://www.nasa.gov/sites/default/files/482462main_1520_-_Independent_Validation_of_Software_Safety_Requirements_for_Systems_of_Systems_Preso-Rev_13.pdf [4] https://s-isac.org/2425-2/ [5] https://www.esa.int/Enabling_Support/Preparing_for_the_Future/Discovery_and_Preparation/Artificial_intelligence_in_space [6] https://www.springboard.com/blog/data-science/ai-space-exploration/ [7] https://standards.nasa.gov/sites/default/files/standards/NASA/B/0/NASA-STD-87398-Revision-B_1.pdf [8] https://news.satnews.com/2023/08/15/space-isac-ai-ml-community-publishes-mlsecops-whitepaper/ [9] https://www.engineering.com/story/keeping-astronauts-safe-with-cloud-based-ai-in-space [10] https://www.linkedin.com/pulse/artificial-intelligence-ai-rockets-omspace [11] https://core.ac.uk/download/pdf/10536783.pdf [12] https://www.whitehouse.gov/briefing-room/statements-releases/2023/09/12/fact-sheet-biden-harris-administration-secures-voluntary-commitments-from-eight-additional-artificial-intelligence-companies-to-manage-the-risks-posed-by-ai/ [13] https://www.forbes.com/sites/tonybradley/2022/04/11/microsoft-and-nasa-leverage-ai-to-streamline-safety-in-space/?sh=24a2f05e21de [14] https://www.aiplusinfo.com/blog/ai-and-space-exploration/ [15] https://hal.science/hal-02271033/document [16] https://readwrite.com/ai-is-taking-over-space-how-artificial-intelligence-will-change-the-future-of-aerospace/ [17] https://www.thecommunityai.org/ai-blog/ai-on-spacex [18] https://aerospace.org/sites/default/files/2020-10/Bailey%20SPD5_20201010%20V2_formatted.pdf [19] https://newspaceeconomy.ca/2023/03/14/how-can-artificial-intelligence-be-used-in-space-technology/ [20] https://aimagazine.com/technology/three-ways-ai-is-being-used-to-help-with-space-exploration [21] https://www.huffpost.com/entry/what-are-the-safety-regulations-for-spacecraft_b_58c8577be4b05675ee9c5b1d [22] https://fedscoop.com/nasa-lack-of-standardized-ai-definition-creates-ai-and-cyber-risks-watchdog-finds/ [23] https://static.aminer.org/pdf/PDF/000/239/089/deriving_safety_properties_of_critical_software_from_the_system_risk.pdf [24] https://www.aiacceleratorinstitute.com/ai-in-space-exploration/ [25] https://www.smithsonianmag.com/air-space-magazine/practicing-safe-software-180962744/ ## aviation AI and software safety in the aviation industry are becoming increasingly important as technology advances. The industry is adopting AI to improve various aspects of aviation, including aircraft maintenance, route optimization, fuel planning, and safety management. Here are some key aspects of AI safety in aviation: 1. **Monitoring Aircraft Systems**: AI can help monitor aircraft systems, automating safety procedures, and detecting potential issues before they become critical[1]. 2. **Predictive Maintenance**: AI systems with built-in machine learning algorithms can collect and analyze flight data to predict maintenance needs, reducing the time and cost associated with inspections and repairs[4]. 3. **Data Analysis**: AI and machine learning can analyze vast amounts of data from various sources, such as flight data recorders, to identify potential safety issues and improve operational efficiency[6]. 4. **Safety Management Systems**: AI-powered safety management systems, such as GE Digital's Safety Insight, help increase flight safety, reduce risk, and save costs by analyzing flight data and providing insights into safety and areas of potential risk[17]. 5. **Regulations and Standards**: Regulatory bodies like the European Union Aviation Safety Agency (EASA) are working on AI roadmaps and guidelines to ensure the safe integration of AI in aviation[3][9]. 6. **Collaboration**: The aviation industry is engaging in collaborative efforts to develop AI safety standards and share best practices, with representatives from firms, regulatory bodies, and academia participating in the process[7]. While AI offers numerous benefits for aviation safety, it is essential to be aware of the risks, such as potential AI malfunction, hacking, and job loss. By adopting a holistic approach to AI implementation and considering both the benefits and risks, the aviation industry can ensure that AI enhances safety while maintaining human involvement and oversight[14]. Citations: [1] https://ts2.space/en/the-impact-of-artificial-intelligence-on-aviation-safety-and-security/ [2] https://aviationweek.com/mro/emerging-technologies/safety-measures-secure-field-loadable-software [3] https://www.easa.europa.eu/en/domains/research-innovation/ai [4] https://www.altexsoft.com/blog/engineering/ai-airlines/ [5] https://www.quora.com/What-measures-are-being-taken-to-ensure-the-safety-and-reliability-of-aircraft-computer-systems [6] https://ntrs.nasa.gov/api/citations/20210018418/downloads/FSF%20ML%20AI%20for%20In-Time%20Aviation%20Safety-final.pdf [7] https://cltc.berkeley.edu/publication/new-report-the-flight-to-safety-critical-ai-lessons-in-ai-safety-from-the-aviation-industry/ [8] https://www.faa.gov/travelers/fly_safe/information [9] https://flightsafety.org/asw-article/safely-implementing-ai/ [10] https://interactive.aviationtoday.com/avionicsmagazine/may-june-2022/ai-in-the-sky-how-artificial-intelligence-and-aviation-are-working-together/ [11] https://www.magellantv.com/articles/aviation-software-is-safety-flying-2nd-class [12] https://www.ainonline.com/aviation-news/blogs/ainsight-now-everywhere-can-ai-improve-aviation-safety [13] https://www.gao.gov/products/gao-21-86 [14] https://www.climaxaviation.com/post/flying-high-with-ai-the-benefits-and-risks-of-artificial-intelligence-in-the-aviation-industry [15] https://www.linkedin.com/pulse/ensuring-aviation-safety-best-practices-lessons-majed-al-marzouqi [16] https://www.futureflight.aero/news-article/2023-07-13/beyond-automation-how-artificial-intelligence-transforming-aviation [17] https://www.ge.com/digital/applications/flight-analytics/safety-insight [18] https://www.linkedin.com/pulse/how-artificial-intelligence-shape-aviation-human-factors-rudi-rademan [19] https://en.wikipedia.org/wiki/Avionics_software ## self driving cars someone who worked at waymo told me they chain multiple models (ai or heuristics) sequentially to ensure maximum security --- Waymo ensures security in its self-driving cars through a combination of safety measures, cybersecurity practices, and continuous improvement based on real-world data. Some of the key aspects of Waymo's security approach include: 1. **Safety Framework**: Waymo has developed a safety framework that evaluates the performance of its autonomous driving system based on three primary capabilities: avoiding incidents with other vehicles or road users, successfully completing trips in fully autonomous mode, and adhering to applicable driving rules where the vehicle is operating[4]. 2. **Cybersecurity**: Waymo has implemented robust processes to identify, prioritize, and mitigate cybersecurity threats. The company's security practices are built on the foundation of Google's security processes and are informed by publications like the NHTSA Cybersecurity Guidance[2]. Waymo can identify attempts to impair vehicle security and has diverse mechanisms for noticing anomalous behavior and internal processes for analyzing those occurrences[2]. 3. **Incident Response Procedure**: In case Waymo becomes aware of an indication that someone has attempted to impair its vehicle's security, the company triggers its company-wide incident response procedure, which involves impact assessment, containment, recovery, and remediation[2]. 4. **Real-World Data and Simulation**: Waymo's autonomous vehicles have driven millions of miles on public roads and billions of miles in simulation, providing valuable data to improve the safety and performance of its self-driving cars[10]. 5. **Safety Performance Assessment**: Waymo regularly assesses the safety performance of its autonomous driving system, publishing papers and sharing methodologies to encourage transparency and advance the science of AV safety analysis[9]. 6. **Collaboration with Industry Partners**: Waymo actively collaborates with other companies and organizations in the autonomous vehicle industry to share knowledge, best practices, and safety standards[2]. By implementing these measures, Waymo aims to reduce traffic injuries and fatalities, ensuring a safe and secure experience for passengers in its self-driving cars[1]. Citations: [1] https://waymo.com/safety/ [2] https://ltad.com/about/cybersecurity.html [3] https://waymo.com/blog/2019/08/waymos-fully-self-driving-vehicles-are.html [4] https://waymo.com/blog/2020/10/revealing-our-approach-to-safety.html [5] https://waymo.com [6] https://waymo.com/blog/2022/09/benchmarking-av-safety.html [7] https://www.theverge.com/2023/9/6/23860029/waymo-insurance-injury-claims-autonomous-vehicle-swiss-re [8] https://www.marketingdive.com/ex/mobilemarketer/cms/news/associations/24221.html [9] https://waymo.com/blog/2023/07/the-waymo-driver-is-already-improving.html [10] https://waymo.com/waymo-driver/ [11] https://www.investopedia.com/articles/investing/052014/how-googles-selfdriving-car-will-change-everything.asp [12] https://www.theverge.com/2022/9/29/23377219/waymo-av-safety-study-response-time-crash-avoidance [13] https://www.theverge.com/2023/3/22/23650792/waymo-safety-case-autonomous-vehicles-av-robotaxi [14] https://www.wgu.edu/blog/how-cybersecurity-drives-self-driving-car-adoption1812.html [15] https://downloads.ctfassets.net/sv23gofxcuiz/4gZ7ZUxd4SRj1D1W6z3rpR/2ea16814cdb42f9e8eb34cae4f30b35d/2021-03-waymo-safety-report.pdf [16] https://www.cnbc.com/2022/03/30/waymo-self-driving-experience-mostly-peaceful-and-productive.html [17] https://support.google.com/waymo/answer/9190838?hl=en [18] https://youtube.com/watch?v=ETeQPZocuE8