Mitigating Dark Web Risks: The Role Of AI And Machine Learning
Dipesh is Group VP at cybersecurity firm Cyble Inc., specializing in the monitoring and mitigation of cybersecurity threats.
gettyThe dark web is a risky and evil realm where unethical and criminal activities flourish. It takes specialized software to access the dark web, which is concealed and inaccessible through standard search engines. It has developed into a well-liked venue for a variety of unlawful activities, including the distribution of malware and hacking tools, the selling of illicit commodities, and the use of software flaws to gain access to sensitive data and systems.
Additionally, it is the location of a number of forums and communities that support and encourage hate speech, extreme ideas and other wrongdoings. The sale of illegal goods like narcotics, weapons and stolen data is one of the most well-known unlawful operations on the dark web. Moreover, the dissemination of viruses and hacking instruments.
Throughout the years, several significant events have impacted the dark web and online crime, including the Silk Road Shutdown, AlphaBay Hack and WannaCry ransomware attack, which demonstrated the ability of cybercriminals to leverage dark web resources. Silk Road was a major darknet marketplace, AlphaBay was shut down and WannaCry had a significant impact on the global computer system.
Cybersecurity professionals and law enforcement must stay vigilant and use emerging technologies to mitigate the risks of the dark web, such as artificial intelligence (AI) and Machine Learning (ML).
AI and ML provide a variety of powerful tools and techniques that can be utilized by law enforcement and cybersecurity professionals to more effectively monitor and address the threats posed by the dark web. For instance, these technologies can be used for threat intelligence where they can analyze vast amounts of data from the dark web to identify patterns and trends in criminal activity. This information can then be used to inform law enforcement actions and develop more effective cybersecurity measures.
Furthermore, AI and ML can also be used for real-time threat detection, helping to identify malicious activities such as the spread of malware or the sale of stolen data, and enabling professionals to take swift action to mitigate these risks. Fraud detection is another area where these technologies excel, as they can analyze large amounts of data to detect and prevent fraudulent activity such as the sale of counterfeit products or the dissemination of false identities.
In addition, sentiment analysis using AI tools can be utilized to identify potential threats by analyzing the language used in dark web forums and other online communities to determine the tone and sentiment of discussions. This information can then be used to inform law enforcement actions.
Finally, AI and ML can be used for predictive analytics, analyzing historical darknet data to predict future trends and activity. This information can help law enforcement officials and cybersecurity professionals anticipate future threats and take proactive steps to mitigate those risks.
Overall, AI and ML are critical tools for managing the risks associated with the dark web, and their use is becoming increasingly important in the fight against cybercrime. In order to safeguard themselves, security leaders can employ a range of tools.
Dark Web ScannersCompanies should identify business-critical information that needs to be protected before implementing a dark web scanning tool. They should research and select a dark web scanning tool that best meets their needs. They should define the scope of the tool, set up the scanner, monitor and analyze results, take action if a threat or breach is detected, evaluate and adjust the scope and response plan, and monitor the evolving threat landscape.
Daily, millions of compromised user records, credit card numbers, intellectual property and login credentials are added to the dark web by cybercriminals. Threat actors utilize the dark web to plan and execute social engineering attacks against key individuals and executives in organizations. Early breach detection empowers security teams to minimize risk exposure, maintain physical security and minimize harm from attacks.
Threat Intelligence PlatformsBefore deploying a threat intelligence platform (TIP), businesses should determine the sorts of threats that must be monitored. They should conduct research and choose the best TIP for their purposes, define the scope, implement the TIP, monitor and analyze results, take action, review and adjust, and monitor the developing threat landscape.
To stay ahead of possible risks, businesses can also monitor the developing threat landscape and alter the TIP settings. With TIP, you can analyze threat actor tactics, techniques, and procedures (TTPs) and redefine your security infrastructure accordingly.
Deep Web Analysis ToolsBusinesses should establish the types of data that need to be reviewed before adopting a deep web analysis tool. They should perform research, select the best tool for their purposes, define the project's parameters, set up the tool, assess the results, act, and then evaluate and tweak the tool. Account creation, setting up the program to crawl specific websites or forums, and integrating it with existing data analysis tools are all examples.
Fraud Detection SoftwareCyber leaders should determine the sorts of fraud that are pertinent to their industry. They ought to do their homework and pick the fraud detection program that best suits their requirements and establish the software's scope, install it, monitor and analyze the results, take prompt action to lessen the damage, and then assess the situation and modify the scope and reaction strategy as necessary. In order to stay ahead of prospective fraud schemes, businesses should also keep an eye on the changing fraud scene and modify the software settings.
AI and machine learning are essential for law enforcement and cybersecurity professionals to monitor and mitigate the risks associated with the dark web, and individuals should exercise caution when accessing it. These technologies can make a critical difference in creating a safe and secure online environment.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
Warren Buffett Loves These AI Stocks, And You Should, Too
© Provided by The Motley Fool Warren Buffett Loves These AI Stocks, and You Should, TooWarren Buffett's Berkshire Hathaway invests in numerous artificial intelligence (AI) stocks. From its vast Apple holdings to non-tech companies such as Chevron, AI has an understated but positive effect on these companies and on the Berkshire portfolio.
CONSTELLATION BRANDS, INC.
Buffett and his team have also ventured into some high-growth AI stocks that could potentially bring outsized returns to shareholders. To this end, investors might want to take a closer look at two Berkshire-owned companies: Snowflake (NYSE: SNOW) and Ally Financial (NYSE: ALLY).
SnowflakeSnowflake has established a leadership position in the data cloud space. Thanks to its software, entities can store data, control updates, and apply their data as they see fit from one centralized, secure location. They can also accomplish this regardless of the cloud-infrastructure environment their data was gathered in, giving Snowflake a competitive edge over competing products from Amazon or Microsoft.
Admittedly, Buffett's team was probably thinking more about the data cloud's potential than AI when it acquired a pre-IPO position in Snowflake in 2020. Nonetheless, AI plays a critical role in supporting Snowflake, specifically with its Data Science & ML (machine learning) application. With the product's centralized data storage, Snowflake can help accelerate ML workflows and grant rapid access to a customer's data.
Also, through its Snowpark developer framework, the company can apply programming languages such as Python and SQL to perform data transformations. This approach can help clients create new ML-driven business insights.
These capabilities appear to increase interest in Snowflake's products. In the fourth quarter of fiscal 2023 (ended Jan. 31), revenue increased 54% year over year. This occurred as Snowflake increased its customer count over that period by 32% to 7,828.
Indeed, quarterly losses increased to $207 million as operating expenses continued to exceed revenue. Still, one could also argue now is the time to buy. Snowflake stock sells at a discount of about 55% from its November 2021 all-time high even after a 20% increase in the stock price over the last year.
While a price-to-sales (P/S) ratio of 27 may seem outlandish in today's tech environment, investors need to remember that Snowflake's sales multiple has never fallen below 20. The P/S ratio also reached a record 184 in late 2020, indicating how much the market has discounted that valuation. Hence, even with its sales multiple, Snowflake offers a compelling value proposition when considering its rapid growth and unique position within the data cloud
Ally FinancialAt first glance, one might wonder how a company focused on loans and personal finance has much need for artificial intelligence. However, investors can find numerous examples of AI in finance, especially with a digital bank like Ally Financial. Ally experienced a significant metamorphosis when the former auto-lending arm of General Motors transformed itself into a digital bank.
Its branchless approach means almost all customer interactions occur online, and its dependence on technology arguably makes AI a necessity for the digital bank. Ally's former chief strategy and corporate development officer once described to American Banker one of its critical AI applications.
By using AI, Ally uses information gathered from other documents to help customers fill out auto-lending forms. This improves the customer experience while also increasing data accuracy. AI also powers a virtual assistant called Ally Assist to answer routine customer service questions without direct human involvement.
Virtual assistants do not seem to discourage customers from banking with or investing in Ally. Its $154 billion in total deposits in 2023's Q1 rose 11% year over year.
But despite that improvement, revenue for 2023's Q1 dropped by 2% year over year. Also, net income for the quarter came in at $319 million, down from $655 million one year ago. Profit fell as Ally increased its provision for credit losses to $446 million, up from $167 million in 2022's Q1.
Admittedly, Buffett's team trimmed its position slightly in that environment, reducing share counts by 3%. Nonetheless, with a 9.6% stake amounting to about 29 million shares, investors should not doubt Berkshire's commitment to or affinity with this stock.
Moreover, with its $1.20 per-share annual dividend, shareholders earn a 4.7% cash return. That likely return on top of potential stock-price growth should bolster confidence in the stock. Furthermore, even with the lower earnings, the price-to-earnings (P/E) ratio stands at 6, a level off its recent lows but down from 17 times earnings in early 2021. At such levels, investors can buy an AI stock inexpensively while earning a significant cash return, increasing the appeal of Ally's value proposition.
SPONSORED:
10 stocks we like better than Snowflake
When our analyst team has a stock tip, it can pay to listen. After all, the newsletter they have run for over a decade, Motley Fool Stock Advisor, has tripled the market.*
They just revealed what they believe are the ten best stocks for investors to buy right now... And Snowflake wasn't one of them! That's right -- they think these 10 stocks are even better buys.
See the 10 stocks
*Stock Advisor returns as of May 15, 2023
Ally is an advertising partner of The Ascent, a Motley Fool company. John Mackey, former CEO of Whole Foods Market, an Amazon subsidiary, is a member of The Motley Fool's board of directors. Will Healy has positions in Berkshire Hathaway. The Motley Fool has positions in and recommends Amazon.Com, Apple, Berkshire Hathaway, Microsoft, and Snowflake. The Motley Fool recommends General Motors and recommends the following options: long January 2025 $25 calls on General Motors. The Motley Fool has a disclosure policy.
Meet The 12 Execs Leading The AI Strategy At Top US Banks Like Goldman Sachs, JPMorgan, And Morgan Stanley
Bank of America
(Clockwise from top left) Bank of America's Hari Gopalkrishnan, Christian Kitchell, Robert Pascal, Teron Douglas, and Nikki Katz. Bank of AmericaKey people:
At Bank of America, AI is organized along business lines.
Chief experience officers are embedded into different business divisions and report into David Tyrie, who serves in a dual role as chief digital officer and chief marketing officer. The CXOs oversee marketing, digital, and data efforts, including how AI is used among employees and customers.
Pascal is the CXO for business, which spans small business, commercial, and investment banking. A McKinsey alum, Pascal also leads Banker Assist, an AI tool that helps investment banking employees with various tasks, such as compiling a client profile ahead of a meeting.
Kitchell, who originally created Bank of America's AI-powered virtual assistant Erica, serves as CXO for wealth. Douglas, who spent 16 years at E*Trade before joining BofA in 2015, covers AI, digital, and data initiatives for consumer.
Meanwhile, Gopalkrishnan and Katz work across business lines, operating as a team in the bank's so-called "hand and glove" structure, which pairs up digital strategy execs with back-end tech experts. Katz reports into Tyrie and Gopalkrishnan reports to Aditya Bhasin, chief technology and information officer.
Katz, who represents digital strategy, works with Gopalkrishnan, whose domain is tech and ops. Combined, they are responsible for how digital products, including the bank's AI-powered chatbot Erica, are deployed.
Citigroup
Prag Sharma, global head of AI and AI center of excellence at Citi. CitigroupKey people:
Sharma, Citi's AI lead, breaks generative AI use cases into two buckets: research acceleration (enabling researchers to go out and summarize info on the internet) and content generation.
The latter is broad, and can be anything from generating code to creating marketing material for sales teams. Sharma said he views AI as a "productivity multiplier," especially among Citi's thousands of developers.
"What we are finding today in generative AI specifically is that it 'understands' language. And if something can understand language, it can be used as a starting point for many, many use cases. I think the imagination is the only stopping point here. But the other point is you need to balance that with the risks and controls," said Sharma, who holds a Ph.D. In computer vision and pattern recognition, which are specializations of AI.
He said there's "strong enthusiasm" from every corner of the bank regarding AI, be it private banking and wealth management, technology and operations, or the institutional businesses, like treasury and trade solutions, markets, and the commercial bank.
All of that interest is funneled through Citi's AI center of excellence, which is run by Sharma. Sharma reports into Nimrod Barak, head of Citi's Innovation Labs where much of the bank's emerging technology research and experimentation takes place.
The purpose of the center of excellence is to ensure there are best practices in place for using AI and machine learning, from governance to risk and tech infrastructure and tooling. Every sector and function of the bank is represented via AI and ML experts who meet on a regular basis to debate everything AI/ML, from which projects to greenlight to firmwide minimum requirements.
Goldman Sachs
Dinesh Gupta, head of business platform engineering and AI engineering, Goldman Sachs. Goldman SachsKey people:
Gupta is a long time veteran of Goldman Sachs, originally coming to the bank in 1996 as a data engineer. He was named a partner in 2022.
Today, he leads AI engineering across the bank, in addition to business platforms engineering, which encompasses various systems, such as Goldman's core payments platform and its trading and portfolio management platform.
He reports into Chief Technology Officer Atte Lahtiranta.
Goldman has a centralized team of natural language specialists and LLM engineers embedded in certain businesses who work with front-office stakeholders to explore potential use cases.
"We are seeing serious momentum within Goldman Sachs and potential use cases are pouring in. We are in the process of prioritizing which use cases we implement first. We've launched several proof of concepts," Gupta told Insider.
One promising use case has been around making software developers more productive, and streamlining and automating away some of the manual work involved in document-heavy operational processes.
"Within Goldman Sachs, if you look at it, we actually get millions of documents every month. To take those documents and apply generative technologies to extract both information from them as well as classify them is an amazing opportunity for us to automate things," Gupta said.
All of the bank's AI projects are mindful of keeping humans in the loop, with the AI acting as a co-pilot, he added.
JPMorgan
(From left to right) JPMorgan's Manuela Veloso and David Castillo. JPMorganKey people:
At the nation's biggest bank, AI is broken into two camps: prospective and current use cases.
Veloso, JPMorgan's head of AI research, leads a task force of academics, who specialize in everything from cryptography to mathematics and electrical engineering, to explore the potential of AI at JPMorgan. Veloso is an academic herself, currently on leave from her post as the head of Carnegie Mellon's machine-learning department.
Unlike most technologists on Wall Street, Veloso's work is not dictated by business stakeholders and her work doesn't always result in a new tool or tech deployment. Rather, Veloso is more focused on exploration, thinking through 'what if' scenarios, and helping others understand, and even embrace, AI.
The team's research is split into seven main areas, from how AI can be used to fight financial crime and help with data management, to using AI to improve how employees work.
The AI research team also founded and manages the bank's Explainable AI Center of Excellence, which was created in July 2020 to share AI techniques and tools and ensure the tech is used fairly.
When potential use cases are identified by Veloso's team, that's where Castillo comes in.
As JPMorgan's firmwide head of AI/ML technology, Castillo oversees the implementation of AI to solve industry use cases. Castillo leads AI efforts across operations, call centers, fraud detection, marketing and advertising, real-time payments and decisioning, and conversational commerce, to name a few, according to his LinkedIn. Castillo's team builds the AI infrastructure, model processes, and controls environment. Drew Cukor, head of AI/ML transformation and engagement and firmwide chief data officer, and his team then use those tools to build models and solutions.
Castillo is also involved in JPM's quantum computing efforts, which are intimately tied to the bank's AI tech stack. He has held tech and AI leadership roles at Boeing, Early Warning Services, and Capital One.
Morgan Stanley
(From left to right) Morgan Stanley's Sreekar Bhaviripudi and Yuriy Nevmyvaka Morgan StanleyKey people:
At Morgan Stanley, there are two key AI executives to know.
Nevmyvaka heads up Morgan Stanley's machine learning research department and is part of the firmwide global innovation team. He holds a Ph.D. From Carnegie Mellon University specializing in computer science and finance, according to his LinkedIn. Nevmyvaka reports into Peter Akwaboah, chief operating officer of technology and head of innovation at Morgan Stanley.
Meanwhile, Bhaviripudi joined the bank in January 2022 to lead AI for the firm's massive wealth business. He is part of Jeff McMillan's team, which oversees analytics, data, and innovation for the wealth management division.
Before Morgan Stanley, Bhaviripudi spent four years at Amazon leading a team of more than 35 machine-learning scientists, engineers, computational linguists, and technical program managers in the Alexa AI natural understanding team, according to his LinkedIn. At Morgan Stanley, he oversees applied AI for things like client personalization and asset- and client-attrition solutions.
The bank is also piloting an AI tool from the same developer as ChatGPT to help advisors parse through Morgan Stanley's trove of research and data, according to a CNBC report. The bank is currently testing it among 900 advisors, with the goal to roll out the tool to its 16,000 advisors in the coming months.
Wells Fargo
Amrith Kumar, head of AI/ML engineering, Wells Fargo. Wells FargoKey people:
Kumar is Wells Fargo's boots-on-the-ground AI and machine learning expert and engineer. He owns the bank's AI and ML platforms, services, tools, and frameworks. Additionally, he is responsible for delivering data engineering and oversees the cloud migration of Wells' AI and ML solutions.
Kumar is a 12-year veteran of Wells Fargo. He worked his way up from a technology manager who worked on data-related services in the wholesale banking division to complete ownership of the bank's AI and ML engineering, according to his LinkedIn.
Kumar reports to Swarup Pogalur, senior vice president and CTO of digital engineering, who reports to group chief information officer, Chintan Mehta.
Some AI use cases at Wells include document analysis and the bank's virtual assistant, Fargo.
