Connect with us

Artificial Intelligence

Security threats in AIs such as ChatGPT revealed by researchers

Published

on

  • University of Sheffield scientists have discovered natural language processing tools (NLP), such as ChatGPT, can be tricked into producing malicious code that could lead to cyber attacks
  • Study is the first to demonstrate that NLP models can be exploited to attack real-world computer systems used in a wide range of industries
  • Results show AI language models are vulnerable to simple backdoor attacks, such as planting a Trojan Horse, that could be triggered at any time to steal information or bring down services
  • Findings also highlight the security risks in how people are using AI tools to learn programming languages to interact with databases

Newswise — Artificial intelligence (AI) tools such as ChatGPT can be tricked into producing malicious code, which could be used to launch cyber attacks, according to research from the University of Sheffield. 

The study, by academics from the University’s Department of Computer Science, is the first to demonstrate that Text-to-SQL systems – AI that enables people to search databases by asking questions in plain language and are used throughout a wide range of industries – can be exploited to attack computer systems in the real world. 

Findings from the research have revealed how the AIs can be manipulated to help steal sensitive personal information, tamper with or destroy databases, or bring down services through Denial-of-Service attacks. 

As part of the study, the Sheffield academics found security vulnerabilities in six commercial AI tools and successfully attacked each one. 

The AI tools they studied were: 

  • BAIDU-UNIT – a leading Chinese intelligent dialogue platform adopted by high-profile clients in many industries, including e-commerce, banking, journalism, telecommunications, automobile and civil aviation
  • ChatGPT
  • AI2SQL
  • AIHELPERBOT
  • Text2SQL
  • ToolSKE 

The researchers found that if they asked each of the AIs specific questions, they produced malicious code. Once executed, the code would leak confidential database information, interrupt a database’s normal service, or even destroy it. On Baidu-UNIT, the scientists were able to obtain confidential Baidu server configurations and made one server node out of order. 

Xutan Peng, a PhD student at the University of Sheffield, who co-led the research, said: “In reality many companies are simply not aware of these types of threats and due to the complexity of chatbots, even within the community, there are things that are not fully understood. 

“At the moment, ChatGPT is receiving a lot of attention. It’s a standalone system, so the risks to the service itself are minimal, but what we found is that it can be tricked into producing malicious code that can do serious harm to other services.”

Findings from the study also highlight the dangers in how people are using AI to learn programming languages, so they can interact with databases.

Advertisement

Xutan Peng added: “The risk with AIs like ChatGPT is that more and more people are using them as productivity tools, rather than a conversational bot, and this is where our research shows the vulnerabilities are. For example, a nurse could ask ChatGPT to write an SQL command so that they can interact with a database, such as one that stores clinical records. As shown in our study, the SQL code produced by ChatGPT in many cases can be harmful to a database, so the nurse in this scenario may cause serious data management faults without even receiving a warning.”

As part of the study, the Sheffield team also discovered it’s possible to launch simple backdoor attacks, such as planting a “Trojan Horse” in Text-to-SQL models by poisoning the training data. Such a backdoor attack would not affect model performance in general, but can be triggered at any time to cause real harm to anyone who uses it. 

Dr Mark Stevenson, a Senior Lecturer in the Natural Language Processing research group at the University of Sheffield, said: “Users of Text-to-SQL systems should be aware of the potential risks highlighted in this work. Large language models, like those used in Text-to-SQL systems, are extremely powerful but their behaviour is complex and can be difficult to predict. At the University of Sheffield we are currently working to better understand these models and allow their full potential to be safely realised.” 

The Sheffield researchers presented their paper at ISSRE – a major academic and industry conference for software engineering earlier this month (10 October 2023). They are working with stakeholders across the cybersecurity community to address the vulnerabilities, as Text-to-SQL systems continue to be more widely used throughout society.

Their work has already been recognised by Baidu whose Security Response Centre officially rated the vulnerabilities as ‘Highly Dangerous’. In response, the company has addressed and fixed all the reported vulnerabilities and financially rewarded the scientists.

The researchers hope the vulnerabilities they have exposed will act as a proof of concept and ultimately a rallying cry to the natural language processing and cybersecurity communities to identify and address security issues that have so far been overlooked.

Xutan Peng added: “Our efforts are being recognised by industry and they are following our advice to fix these security flaws. However, we are opening a door on an endless road – what we now need to see are large groups of researchers creating and testing patches to minimise security risks through open source communities.

Advertisement

“There will always be more advanced strategies being developed by attackers, which means security strategies must keep pace. To do so we need a new community to fight these next generation attacks.”

Journal Link: The 34th IEEE International Symposium on Software Reliability Engineering

Source: University of Sheffield

Author


Discover more from Daily News

Subscribe to get the latest posts sent to your email.

Science

Nauticus Robotics Enters Aquanaut Mark 2 Testing Partnership with FAU

Published

on

Nauticus Robotics
Stock image of an underwater robot.

HOUSTON /PRNewswire/ — Nauticus Robotics, Inc. (NASDAQ: KITT), a leading innovator in subsea robotics and software, announces a partnership with Florida Atlantic University (FAU) for further testing of its flagship underwater vehicle, Aquanaut Mark 2.

Nauticus has contracted with FAU to perform Aquanaut Mark 2 vehicle development testing in the protected waters near the university’s Institute for Ocean and Systems Engineering Dania Beach facilities as well as in the deeper waters of the Atlantic Ocean. This partnership is expected to result in approximately 90% savings on vehicle development testing. This phase of testing requires both shallow and deeper waters to fully test functionality while maintaining real-time communication via tether for instant feedback. It also provides the opportunity to carry out tetherless missions that simulate operations the Aquanaut vehicle is designed to accomplish. Without this partnership, testing in mid-range water depths would require the use of more expensive ocean-going vessels.

“Working at FAU this summer has provided a perfect platform to efficiently test the Aquanaut vehicle, in both protected waters for product development and open waters for data collection and sea trials. The location, and campus, at FAU has allowed us to expedite the Aquanaut system technology in preparation for commercial use,” stated Daniel Dehart, Nauticus Robotics’ Vice President of Field Operations.

Nauticus plans to continue testing Aquanaut vehicles in conjunction with FAU. The Aquanaut vehicle currently undergoing its final testing will move to the Gulf of Mexico later this month to begin commercial operations. Once ready, the second Aquanaut vehicle is planned to take its place at FAU to continue development of the next round of capabilities.

The Aquanaut vehicle leverages Nauticus’ proprietary ToolKITT software, a versatile and platform-independent solution previously tested across various subsea vehicle classes. By harnessing the power of ToolKITT and the pioneering design of Aquanaut, Nauticus is spearheading the industry’s shift from traditional tethered operations to augmented autonomy. This initiative aims to fully transform subsea operations into autonomous, highly efficient processes.

About Nauticus Robotics

Nauticus Robotics, Inc. develops autonomous robots for the ocean industries. Autonomy requires the extensive use of sensors, artificial intelligence, and effective algorithms for perception and decision allowing the robot to adapt to changing environments. The company’s business model includes using robotic systems for service, selling vehicles and components, and licensing of related software to both the commercial and defense business sectors. Nauticus has designed and is currently testing and certifying a new generation of vehicles to reduce operational cost and gather data to maintain and operate a wide variety of subsea infrastructure. Besides a standalone service offering and forward-facing products, Nauticus’ approach to ocean robotics has also resulted in the development of a range of technology products for retrofit/upgrading traditional ROV operations and other third-party vehicle platforms. Nauticus’ services provide customers with the necessary data collection, analytics, and subsea manipulation capabilities to support and maintain assets while reducing their operational footprint, operating cost, and greenhouse gas emissions, to improve offshore health, safety, and environmental exposure. 

SOURCE Nauticus Robotics, Inc.

https://prnmedia.prnewswire.com/news-releases/nauticus-robotics-enters-aquanaut-mark-2-testing-partnership-with-fau-302218514.html

Advertisement

The science section of our news blog STM Daily News provides readers with captivating and up-to-date information on the latest scientific discoveries, breakthroughs, and innovations across various fields. We offer engaging and accessible content, ensuring that readers with different levels of scientific knowledge can stay informed. Whether it’s exploring advancements in medicine, astronomy, technology, or environmental sciences, our science section strives to shed light on the intriguing world of scientific exploration and its profound impact on our daily lives. From thought-provoking articles to informative interviews with experts in the field, STM Daily News Science offers a harmonious blend of factual reporting, analysis, and exploration, making it a go-to source for science enthusiasts and curious minds alike. https://stmdailynews.com/category/science/

Author


Discover more from Daily News

Subscribe to get the latest posts sent to your email.

Continue Reading

Tech

The Weather Network launches its first advertising campaign using an AI-assisted Avatar

Published

on

OAKVILLE, ON, June 11, 2024 /CNW/ – The Weather Network has launched a first-of-kind ad campaign using an AI-Assisted Avatar of their own Content Creator and Presenter, Rachel Schoutsen. The Weather Network partnered with Home Hardware Stores Limited to showcase storefront and local store callouts where Rachel’s AI-Assisted Avatar covers an Ontario summer forecast, plus a summer sale call to action, designed to drive traffic to Home Hardware locations across Ontario.

Watch this video to see Rachel Schoutsen’s avatar in action

Rachel Schoutsen’s avatar in action.

By leveraging advanced AI technologies, The Weather Network is now able to deliver hyper-local forecasts on a larger scale, providing communities and local economies with opportunities to prepare for and respond to weather-related events. The Weather Network’s mission is to keep its users safe and informed with accurate and localized weather information.

“This partnership puts Home Hardware front and centre when Canadians are engaging with weather content and planning for the summer ahead,” said Simon Jennings, CRO, The Weather Network. “Canadians want to know what to expect as they plan their busiest social days and holidays and can purchase the right products for the weather forecast in their area.”

“Home Hardware is a business built on serving small communities across the country and we are proud to support The Weather Network’s innovative efforts to do the same through AI technology,” said Laura Baker, Chief Marketing Officer, Home Hardware Stores Limited. “Our more than 1,000 stores and the Dealers who operate them make it their mission to provide comprehensive support for all our customers’ home improvement needs within their communities. That means helping Canadians complete and enjoy their outdoor home improvement and yard maintenance projects while also preparing for any inclement weather events.”

With a focus on helping users and businesses make informed decisions using cutting-edge technology, The Weather Network is committed to ensuring the use of avatars in its promotional materials aligns with the highest ethical standards. All promotional materials have a clear disclaimer to ensure its viewers are well informed of AI-created content. In addition, employee avatars are used with their consent and an agreement to participate.

“We are excited about this latest innovation and the opportunities it will provide to our consumers and businesses,” said Maureen Rogers, Managing Director Editorial & Public Alerting, The Weather Network. “As with any new product we launch, we value their input and will use it to continuously improve the product to meet their needs and deliver products that exceed their expectations. We are delighted to have partnered with Home Hardware on piloting this new product with us.”

Advertisement

The Weather Network recently launched Your Weather Assistant, a generative AI tool to transform the way people consume weather information. The introduction of its newly developed AI-Assisted Avatar demonstrates Pelmorex Corp.’s, owner of The Weather Network, commitment to innovation as it evolves into an AI-frontiered weather and climate organization.

About Pelmorex Corp.
Pelmorex Corp., founded in 1989, is an international weather information and data management company. Pelmorex owns and operates the weather brands The Weather Network, MétéoMédia, Eltiempo.es, Clima, and Otempo.pt. It also operates Canada’s National Alert Aggregation and Dissemination System, part of Alert Ready. Through constant innovation and entrepreneurship, Pelmorex has grown to reach consumers around the globe, has become one of the largest weather information providers and has broken new ground in providing data solutions and insights to businesses. Through harnessing the value of weather, Pelmorex is driven to make the world smarter and safer for consumers and businesses. 

To learn more, visit pelmorex.com and follow us on X (platform formerly known as Twitter), Instagram and LinkedIn.

About Home Hardware Stores Limited

Founded 60 years ago in St. Jacobs, Ontario, Home Hardware Stores Limited is proudly Canadian and the country’s largest Dealer-owned and operated home improvement retailer with more than 1,000 stores operating under the Home Hardware, Home Building Centre, Home Hardware Building Centre and Home Furniture banners. The Home Hardware Dealer network improves life at home by providing superior home improvement retail experiences through helpful advice and a deep understanding of the communities they serve. Home Hardware Stores Limited is designated as one of Canada’s Best Managed Companies and recognized as one of Canada’s Best Employers. More information about the company is available at homehardware.ca.

SOURCE Pelmorex Corp.

https://stmdailynews.com/category/tech

Advertisement

Author


Discover more from Daily News

Subscribe to get the latest posts sent to your email.

Continue Reading

Artificial Intelligence

AI Marketing Strategies On Generic AI Platforms Set To Hinder Success

Businesses overusing AI will find themselves disadvantaged when developing marketing strategies and marketing programs in 2024 says Robotic Marketer founder.

Published

on

ATLANTA, GEORGIA, UNITED STATES /EINPresswire.com/ — As the business world stands on the precipice of 2024, marketing industry expert, Mellissah Smith, forecasts a seismic shift in the competitive landscape due to the overuse and misuse of generic AI in marketing.

The pivotal new year, marked by the emergence of a stark competitive gap caused by a one-size-fits-all approach provided by many AI platforms, heralds a call to action for businesses worldwide that are being led by over-use of AI for content and strategy without bespoke data analysis and benchmarking.


AI is a productivity tool. It’s a tool to do fast analysis. Without using AI tools that integrate with your tech stack, it is a one-size-fits all, and something not to be relied upon.”
— Mellissah Smith


“In the race to embrace AI, the market is inadvertently veering towards a divide that will separate the leaders from the laggards,” says Mellissah Smith, the founder behind Robotic Marketer, a world-first AI powered marketing strategy technology platform. “2024 will witness the dawn of an AI divide where businesses that recognize and act upon the need for personalized, data driven AI strategies will gain a distinct competitive edge. Those continuing down the path of generic AI solutions will find themselves outpaced and outmanoeuvred by companies that are using their available data and content personalization to connect with customers. This isn’t just a minor shift; it’s a critical juncture that will significantly impact the competitive landscape in 2024 and the years that follow.”


AI Marketing Strategy Company Founder, Mellissah Smith from Robotic Marketer

Mellissah Smith, with her three decades of marketing experience, states, “2024 isn’t just another year; it’s 12-months since Open AI’s ChatGPT changed the way everyday people thought about AI and only 6 months since many companies have realized that they can’t afford not to get on the bandwagon of AI. However, the overuse of generic AI has reached its zenith, and this has been largely due to people not knowing how AI is developed and how it will best suit their company needs. AI is now intricately woven into the fabric of each business’s unique narrative, so it’s time to create some maturity in its use.”

Smith addresses the critical shortfall in current AI applications: “The market is inundated with tools that offer a one-dimensional view, leading to strategies that are as broad as they are bland. Our commitment at Robotic Marketer is to transcend this norm and as a company that has been in the market since 2017, we have been able to create the labyrinth of machine learning. We’re introducing an AI that doesn’t just analyze but understands, interprets, and develops strategies with a level of precision and personalization previously unimagined.”

Robotic Marketer’s innovation sets a new precedent in both the marketing and consulting industries. “We’re not just creating another marketing tool; we’re creating a AI platform that becomes the heart of your marketing strategy. It’s about developing a dynamic, intuitive understanding of your business goals, market nuances and customer dynamics,” Smith states.

As businesses prepare to navigate the complexities of 2024, Smith underscores the imperative for a strategic renaissance. “The future is not about broad strokes but about the detail and depth that creates relevance.”

“The economy will dictate how many businesses navigate 2024, which requires for those companies most effected by a downturn to automate tasks where possible. Scaling marketing efforts is possible with AI, particularly if you have the right marketing strategy in place, but must come with personalization.”

Advertisement

About Robotic Marketer:
Founded by the visionary and “Innovator of the Year” recipient, Mellissah Smith, Robotic Marketer is at the vanguard of the next generation of AI-driven marketing strategy development. Committed to innovation, personalization, and strategic acumen, Robotic Marketer is empowering businesses worldwide to navigate the future of marketing with unparalleled precision and insight. Working with brands such as SAP, Oracle, Mitel and Atlassian through channel programs, and 1000’s of companies globally, the company has firmly established its position as the leading AI marketing strategy tool that develops, manages, executes and reports on marketing strategy performance. The platform integrates with social media, Google, competitor insights, GPT-4, and popular email marketing, CRM, marketing automation and advertising tools. The company is expanding globally through marketing agencies using the platform as a productivity tool with inbuilt benchmarking and industry best practice. www.roboticmarketer.com



Contact:
Mellissah Smith
Robotic Marketer
info@roboticmarketer.com
www.roboticmarketer.com


Visit on social media:
Facebook
Twitter
LinkedIn
Instagram
YouTube

AI marketing strategy technology platform Robotic Marketer

Source: Robotic Marketer

Author


Discover more from Daily News

Subscribe to get the latest posts sent to your email.

Continue Reading

Trending