AI Cert Flashcards

1
Q

Deep Learning is a subset of machine learning which uses layered networks to perform decision making based on either supervised or unsupervised date. What is the name of the network that helps Deep Learning perform this decision making?

A. Neural networks
B. Rule-based systems
C. Predictive analytics

A

Answer - A

Neural Networks - A layered network of connections that are influenced by weights and biases.

A layer is a collection of nodes in a neural network grouped together for specific computations. Layers organize the flow of information, with input and output layers handling data and intermediate hidden layers facilitating complex feature extraction.
A weight is a numerical value assigned to the connection between nodes in a neural network. It represents the strength of influence that one node’s output has on another and adjusted during training to optimize the network’s performance in learning and making predictions.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Which of the following is an example of Deep Learning? Which of the following learning methods would be most similar to how Deep Learning works in this scenario?

A. Memorizing a list of fruit names and their corresponding features
B.Studying a set of photographs of fruits and their labels
C. Looking at examples of fruits and learning to recognize patterns without explicit instruction
D. Asking someone to describe characteristics of each fruit to you

A

Answer - C

A is wrong because Deep Learning involves learning patterns and features from data without explicit programming or instruction.
B s wrong because Deep Learning involves unlabeled data. Labeled data, is characteristic of supervised learning rather than Deep Learning.
D is wrong beacuse Deep Learning involves the model learning patterns and features directly from data without explicit human instructions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is a key characteristic of machine learning in the context of AI capabilities?
A. Can perfectly mimic human intelligence and decision-making
B. Relies on preprogrammed rules to make decisions
C. Utilizes algorithms to learn from data and make decisions

A

Answer - C

Machine Learning is like following a recipe to bake a cake. You have a set of instructions (the recipe) and ingredients, and you use them to make the cake. Similarly, in ML, you provide the computer with a set of instructions (algorithms) and data (ingredients), and it learns from the data to accomplish tasks without being explicitly programmed.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Machine learning is the process of using large amounts of data to train a model to make predictions, instead of handcrafting an algorithm. Machine learning uses _____________ , _____________ , _____________ , and _____________, types of data.

A

Answer - Unstructured/Structured Data and Unsupervised/Supervised Data

Unstructured Data - Example, News Article
Structured Data - Example, Spreadsheet
Unsupervised - Unmonitored data that AI tries to find connections in the data without really knowing what it’s looking for.
Supervised - Monitored data where make sure every piece of input data has a matching, expected output that we can verify.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are the foundational components of AI systems?

A. Algorithms, software, and hardware elements
B. Algorithms, data, and hardware infrastructure
C. Algorithms, data, and computational resources

A

Answer - C

Algorithms - Rules for inputs and outputs
Data - Used to train the AI
Computational Resources - physical and virtual resources

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Match the AI capabilities to their everyday uses.

Numeric Prediction = ?
Classification = ?
Robotic Navigation = ?
Language Processing = ?

Options =
Stock Market
Weather Forecasting
Bank Fraud Alerts
Flagged Comments on Social Media
Phishing Emails
Roomba Vacuum
Hands Free Driving
ChatGPT

A

Numeric Prediction = Stock Market,
Weather Forecasting

Classification = Bank Fraud Alerts,
Flagged Comments on Social Media,
Phishing Emails

Robotic Navigation = Roomba Vacuum, Hands Free Driving

Language Processing = ChatGPT

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

A company wants to reduce the workload of its customer service team by offering a chatbot on its website to handle common queries. Which AI technique is best suited for this situation?

A. Visual data analysis
B. Predictive data insights
C. Natural language processing

A

Answer - C

Natural language processing is a chatbots way of understanding human language. Natural Language is used for chatbots in order to talk to customers and answer their questions. NLU tech helps the chatbot understand what customers say and give them good answers. This helps a lot with handling common questions and making customer support better.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

In the realm of AI capabilities, what is the primary function of computer vision?

A. Analyzing and comprehending visual information
B. Improving images through image processing techniques
C. Forecasting future results using data

A

Answer - A
Computer vision is a type of AI that interprets and understands visual data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What type of AI is being described?
This AI can use Machine learning techniques to produce unique outputs based on a given input.
This AI can create new content, such as text, images, or even music, by learning patterns and relationships in the input data and generating new data that fit those patterns.

A. Generative
B. Predictive
C. Probabilistic

A

Answer - A
Generative AI creates new content based on existing data prompted by humans
Combines algorithms and deep learning neural networks techniques to generate content that is based on the patterns it observes in other content

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What type of AI is being described?
This AI Studies historical data, identifies patterns and makes predictions about the future that can better inform decisions
This AI is used for financial forecasting, fraud detection, healthcare and marketing
A. Generative
B. Predictive
C. Probabilistic

A

Answer - B
Predictive artificial intelligence (AI) refers to the use of machine learning to identify patterns in past events and make predictions about future events

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Match the term to the scenarios

A business is using Generative AI to make art and some of the faces have unrealistic features such as 3 eyes.

A developer is using Generative AI to help code a program is feeding the AI prioteiary data to debug.

A student is using Generative AI to help write a research paper and neglects to properly cite all of the sources used.

A person is using Generative AI to make bot accounts on social media.

A company is auto installing ChatGPT on every device and encourages every employee to make a least 1 prompt a day.

A

A business is using Generative AI to make art and some of the faces have unrealistic features such as 3 eyes.
-Hallucinations

A developer is using Generative AI to help code a program is feeding the AI prioteiary data to debug.
-Data Security

A student is using Generative AI to help write a research paper and neglects to properly cite all of the sources used.
-Plagiarism

A person is using Generative AI to make bot accounts on social media.
-User Spoofing

A company is auto installing ChatGPT on every device and encourages every employee to make a least 1 prompt a day.
-Sustainability

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Which statement best represents a significant obstacle in human-AI cooperation in decision-making?

A. AI facilitates more knowledgeable and impartial decision-making
B. AI diminishes the necessity for human participation in decision-making procedures
C. AI encourages dependence on AI, possibly reducing critical thinking and supervision

A

Answer - C

Over-reliance on AI can potentially lead to less critical thinking and oversight.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

A business analyst (BA) is looking to boost their company’s performance by making improvements in their sales procedures and customer service.
What AI tools could the BA employ to address these requirements?

A. Lead Scoring, Opportunity forecasting, and Case Classification
B. Cleaning up sales data and ensuring proper governance of customer support data.
C. Utilizing machine learning models and predicting chatbot behavior

A

Answer - A
Lead scoring, opportunity forecasting, and case categorization are important AI applications for a business analyst (BA) aiming to enhance their company’s sales processes and customer support. These AI applications help the BA by providing data-driven insights, automating manual tasks, and improving decision-making processes, ultimately leading to improved sales and customer support performance.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Which Einstein capability uses emails to create content for Knowledge articles?

A. Generate
B. Predict
C. Discover

A

Answer - A

Einstein Generate is a natural language generation (NLG) feature that can automatically write summaries, descriptions, or recommendations based on data or text inputs. For example, Einstein Generate can analyze email conversations between agents and customers and generate draft articles for the Knowledge base.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What are some key benefits of AI in improving customer experiences in CRM?

A. Streamlines case management by categorizing and tracking customer support cases, identifying topics, and summarizing case resolutions
B. Fully automates the customer service experience, ensuring seamless automated interactions with customers
C. Improves CRM security protocols, safeguarding sensitive customer data from potential breaches and threats

A

Answer - A

Streamlines case management by categorizing and tracking customer support cases, identifying topics, and summarizing case resolutions. AI in CRM categorizes cases, tracks support types, prioritizes cases, monitors status, identifies topics, reasons, and closure codes, and tracks case types and channels. This leads to more personalized and efficient customer service.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What are three frequently employed examples of AI in CRM?

A. Predictive scoring, forecasting, recommendations
B. Predictive scoring, reporting, image classification
C. Einstein Bots, face recognition, recommendations

A

Answer - A

These are three common uses of AI in CRM involving predicting customer behavior, forecasting future trends, and providing personalized recommendations to enhance customer engagement and sales efficiency within the CRM system.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

How does an organization benefit from using AI to personalize the shopping experience of online customers?

A. Customers are more likely to share personal information with a site that personalizes their experience.
B. Customers are more likely to be satisfied with their shopping experience.
C. Customers are more likely to visit competitor sites that personalize their experience.

A

Answer - B

Using AI to personalize the online shopping experience leads to increased customer satisfaction. This happens because personalized recommendations make shopping more convenient, engaging, and relevant, which ultimately boosts conversion rates, customer retention, and loyalty.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

In what way does AI aid in the process of lead qualification?

A. Generates customized SMS marketing campaigns
B. Engages with potential customers automatically
C. Evaluates leads using customer information

A

Answer - C

AI assists in the lead qualification process by analyzing customer data. AI algorithms can assess various aspects of leads, such as their behavior, demographics, and interactions with a company’s website or products. By processing this information, AI can assign scores or labels to leads, indicating their likelihood to convert into customers. This automated evaluation streamlines the lead qualification process and helps sales teams prioritize their efforts on leads that are more likely to result in successful conversions.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Cloudy Computing wants to use AI to enhance its sales processes and customer support. Which capability should they use?

A. Dashboard of Current Leads and Cases
B. Sales Path and Automated Case Escalations
C. Einstein Lead Scoring and Case Classification

A

Answer - C

Einstein Lead Scoring enhances your sales processes and Case Classification enhances your customer support processes using AI capabilities.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Which features of Einstein enhance sales efficiency and effectiveness?

A. Opportunity Scoring, Opportunity List View, Opportunity Dashboard
B. Opportunity List View, Lead List View, Account List View
C. Opportunity Scoring, Lead Scoring, Account Insights

A

Answer - C

Opportunity Scoring, Lead Scoring, and Account Insights are all features of Einstein that contribute to enhancing sales efficiency and effectiveness.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

A marketing manager wants to use AI to better engage with their customers. Which functionality provides the best solution?

A. Bring Your Own Model
B. Journey Optimization
C. Einstein Engagement

A

Answer - B

  • With Salesforce Marketing Cloud Engagement.

Journey Optimization allows one to Create, test, and optimize personalized campaign variations with built-in predictive AI. Make every moment count by automating and customizing all aspects of customer engagement — including channel, content, timing, and send frequency. Scale dynamic journeys and improve productivity with AI.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

A service leader plans to use AI to help customers resolve their queries more quickly with a guided self-service app. Which Einstein feature offers the most suitable solution for this?

A. Automated Chatbots
B. Categorizing Cases
C. Offer personalized recommendations

A

Answer - A

Chatbots in a self-service app interact with customers in real time, understand their questions, and offer quick solutions, making self-help more efficient and user-friendly.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

A sales manager wants to improve Salesforce operations with AI. What AI application would offer the greatest benefits?

A. Handling and organizing data effectively
B. Generating sales-related dashboards and reports
C. Prioritizing leads and predicting sales opportunities

A

Answer - C

In Salesforce, AI improves sales by scoring leads and predicting future opportunities, helping leaders prioritize leads and enhance sales processes.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What is the primary benefit of using generative AI in CRM for customer support?

A. Generating more marketing emails
B. Reducing the need for human customer support agents
C. Increasing customer wait times

A

Answer - B

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

What salesforce product leverages AI to provide insights and recommendations to sales and service teams?

A. Salesforce Analytics
B. Salesforce Marketing Cloud
C. Salesforce Einstein

A

Answer - C

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What is the purpose of Salesforce Einstein Discovery?

A. To create advanced AI models from scratch
B. To predict customer behavior based on historical data
C. To automate email marketing campaigns

A

Answer - B

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What is the primary function of Einstein Prediction Builder?

A. To forecast future sales opportunities
B. To categorize customer support tickets
C. To design email templates

A

Answer - A

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

Which Salesforce AI application is recommended to enhance sales processes?

A. Einstein Prediction Builder.
B. Einstein Voice
C. Einstein Lead Scoring

A

Answer - C

Einstein Lead Scoring is specifically designed to enhance sales processes by scoring leads based on their likelihood to convert, allowing sales teams to prioritize their efforts effectively.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

Which feature of Marketing Cloud Einstein uses AI to predict consumer engagement with email and MobilePush messaging?

A. Content Selection.
B. Email Recommendations.
C. Engagement Scoring.

A

Answer - C

Customer data and machine learning are used to assign scores for every contact ‘s likelihood to engage with emails and interact with push notifications.

30
Q

Match the scenarios to the terms

Continuous Data
Discrete Data

Options
-Temperature in a room
-Time it takes to run a race
-Stock price fluctuations
-Heigh and weight
-Population growth
-Duration of sunlight
-Kids in a classroom
-Number of books on a bookshelf
-Number of cars passing through a toll booth
-Number of molecules in a chemical reaction

A

Continuous Data
-Temperature in a room
-Time it takes to run a race
-Stock price fluctuations
-Heigh and weight
-Population growth
-Duration of sunlight

Discrete Data
-Kids in a classroom
-Number of books on a bookshelf
-Number of cars passing through a toll booth
-Number of molecules in a chemical reaction

31
Q

Match the term to the scenarios
(Nominal - is not definitively ranked eg favorite fruits - bananas, apples etc. Nominal = Named.
Ordinal - is definitively ranked eg. cases - high, medium, low priority. Ordinal = Ordered.)

Online Survey Analysis - You are analyzing data from an online survey about favorite movie genres. Participants were asked to select their favorite genre from a list of options. What types of data is being collected?

Restaurant Rating - In a recent survey, customers were asked to rate their dining experience at a restaurant on a scale of 1 to 5 stars. What types of data is being collected?

Clothing Sizes - When categorizing clothing sizes as small, medium, large, and extra-large. What type of data classification are you using?

Car Colors - In a survey about car preferences, respondents were asked to choose their favorite car color from a list of options: red, blue, black, white, and green. What types of data is being collected?

A

Online Survey Analysis - You are analyzing data from an online survey about favorite movie genres. Participants were asked to select their favorite genre from a list of options. What types of data is being collected?
-Nominal

Restaurant Rating - In a recent survey, customers were asked to rate their dining experience at a restaurant on a scale of 1 to 5 stars. What types of data is being collected?
-Ordinal

Clothing Sizes - When categorizing clothing sizes as small, medium, large, and extra-large. What type of data classification are you using?
-Ordinal

Car Colors - In a survey about car preferences, respondents were asked to choose their favorite car color from a list of options: red, blue, black, white, and green. What types of data is being collected?
-Nominal

32
Q

What role does data quality play in accomplishing AI business goals?

A. Data quality is essential for generating precise AI data insights
B. Data quality is not needed because AI can handle all types of data
C. Data quality is crucial for adhering to AI data storage constraints

A

Answer - A

A. Data quality is essential for generating precise AI data insights

  • High-quality data is crucial for training AI models, making accurate predictions, and providing valuable insights. Poor-quality data can lead to inaccurate or biased AI results, which can hinder the achievement of business objectives. Therefore, ensuring data quality is a fundamental requirement for AI to deliver meaningful and reliable insights that can inform business decisions and strategies.
33
Q

What effect does a data quality evaluation have on business results for companies utilizing AI?

A. Enhances the efficiency of AI recommendations
B. Speeds up the introduction of new AI solutions
C. Establishes a baseline for AI predictions

A

Answer - C

C. Establishes a baseline for AI predictions

  • Assessing data quality provides insights into the quality of data, which is crucial for ensuring accurate AI outcomes.
34
Q

A company employs Einstein for generating predictions but is experiencing inaccuracies. What could be a possible explanation for this?

A. Low data quality
B. Excessive data volume
C. Incorrect product choice

A

Answer - A

A. Low data quality

  • Good quality data is crucial for accurate predictions. Poor data quality can lead to inaccurate predictions.
35
Q

What is the expected outcome of high-quality data on customer relationships?

A. Improved customer trust and satisfaction
B. Increased brand loyalty
C. Increased expenses for acquiring customers

A

Answer - A

A. Improved customer trust and satisfaction

  • When a business uses high-quality data effectively, it can better understand its customers’ needs and preferences. This enables the company to provide more personalized and relevant experiences, products, and services. As a result, customers tend to trust the brand more and are more satisfied with their interactions, ultimately leading to improved customer trust and satisfaction.
36
Q

A company depends on data analysis to optimize its product recommendations; however, they are encountering a recurring issue of incomplete customer records, with missing contact information and incomplete purchase histories.
How will this incomplete data quality impact the company’s operations?

A. The accuracy of product recommendations is hindered.
B. The diversity of product recommendations is improved.
C. The response time for product recommendations is stalled.

A

Answer - A
A. The accuracy of product recommendations is hindered.

  • Without comprehensive and accurate customer data, the AI system may struggle to make precise recommendations, potentially impacting the company’s ability to provide relevant and effective product suggestions to customers. This incomplete data quality can hinder the accuracy and relevance of the recommendations, which can, in turn, affect the company’s operations and customer satisfaction.
37
Q

A system possesses a significant volume of data, yet it is dispersed across various systems and lacks standardization. Which fundamental data quality aspect should a developer use to prioritize and guarantee the efficiency of their AI models?

A. Age
B. Volume
C. Performance
D. Consistency
E. Completeness

A

Answer - D

D. Consistency

  • It’s important to emphasize the significance of consistency as a fundamental data quality factor. Data volume and data location, on the other hand, are not directly tied to data quality.
38
Q

A company is enhancing their predictive accuracy of its AI model by leveraging a substantial volume of data. What data quality aspect is the company prioritizing?

A. Accuracy
B. Volume
C. Performance
D. Location
E. Completeness

A

Answer - A

A. Accuracy

  • High-quality, accurate data is essential for training AI models that make precise predictions. Inaccurate data can lead to incorrect model outputs and reduced prediction quality. Therefore, ensuring the accuracy of the data is crucial to achieving more reliable and effective AI predictions.
39
Q

A Business Analyst (BA) is in the process of creating a new AI use case. As part of their preparations, they generate a report to examine whether there are any null values in the attributes they intend to utilize. What data quality aspect is the BA confirming by assessing null values?

A. Usage
B. Duplication
C. Completeness
D. Age
E. Volume

A

Answer - C
C. Completeness

  • For each business purpose, make a list of the necessary fields. Afterward, generate a report indicating the percentage of empty values in these fields. Alternatively, you can employ a data quality app from AppExchange.
40
Q

A company intends to employ an AI model for forecasting shoe demand based on historical sales data and regional attributes. Which data quality dimension is crucial for achieving this objective?

A. Reliability
B. Volume
C. Age
D. Duplication
E. Completeness

A

Answer - C
C. Age

  • The Age, Completeness, Accuracy, Consistency, Duplication, and Usage of a dataset are vital factors to assess when determining its suitability for AI models. However, the size and the number of variables in the dataset are unrelated to its appropriateness for AI models.
41
Q

A company wants to ensure that multiple records for the same customer are removed from Salesforce. Which feature should be used to accomplish this?

A. Duplicate management
B. Trigger deletion of old records
C. Standardized field names

A

Answer - A
Duplicate management

  • Duplicate management in Salesforce is a feature that allows you to identify and handle duplicate records effectively. It provides tools to detect and merge duplicate records, ensuring that only a single, accurate record is retained for each customer
42
Q

An administrator wants to ensure that a field is set up on the customer record so their preferred name can be captured. Which Salesforce field type should the administrator use to accomplish this?

A. Rich Text Area
B. Text
C. Multi-Select Picklist
D. Long Text Area

A

Answer - B
B. Text

  • A text field allows for the entry of a single text value, which is appropriate for capturing a customer’s preferred name.
43
Q

A Salesforce administrator creates a new field to capture an order’s destination country. Which field type should they use to ensure data quality?

A. Number
B. Picklist
C. Text
D. Multi-Select Picklist

A

Answer - B
B. Picklist

  • A picklist field allows for the entry of a single value, which is appropriate for capturing a customer’s country with standardized naming conventions
44
Q

During a conversation with a customer considering AI implementation in Salesforce, what should be the consultant’s top priority when discussing the ethical aspects of data management?

A. Privacy, bias, compliance, and security
B. Visualization, data storage, and retrieval
C. Network, software, and hardware

A

Answer - A
The consultant’s main concerns with AI Ethics should be privacy, bias, security, and following the rules. These things make sure AI is used responsibly and that people’s data is treated with respect.

45
Q

What is the term for bias that imposes the values of a system onto others?

A. Societal
B. Automation
C. Association

A

Answer - B

Automation bias means a system forces its own ideas onto others. For example, in a beauty contest judged by AI in 2016, the AI mostly picked white winners because it was trained on pictures of white women and didn’t recognize the beauty in people with different features or skin colors. This shows how the bias in the AI’s training data affected the contest’s results

46
Q

Cloud Kicks implements a new product recommendation feature for its shoppers that recommends shoes of a given color to display to customers based on the color of the products from their purchase history. Which type of bias is most likely to be encountered in this scenario?

A. Societal
B. Confirmation
C. Survivorship

A

Answer - B

Confirmation bias is most likely to be encountered in this scenario. Confirmation bias is a type of bias that occurs when data or information confirms or supports one’s existing beliefs or expectations. For example, confirmation bias can occur when a product recommendation feature only recommends shoes of a given color based on the customer’s purchase history, without considering other factors or preferences that may influence their choice

47
Q

What is the significance of data protection measures in AI usage?

A. Expands the range of data collected
B. Enhances the quality of data
C. Safeguards privacy and compliance

A

Answer - C

Data protection measures are primarily implemented to ensure privacy and compliance with regulations.

48
Q

Why is the explainability of trusted AI systems important

A. Clarifies how AI models reach decisions
B. Adds complexity to AI models
C. Boosts the security and precision of AI models

A

Answer - A

Explainability in AI systems is about providing clear explanations of how AI models make decisions.

49
Q

What purpose do Salesforce’s Trusted AI Principles serve within CRM systems?

A. Establishing a structure for AI data model precision
B. Guiding the ethical and responsible utilization of AI
C. Defining the technical requirements for AI integration

A

Answer - B

Salesforce’s Trusted AI Principles are a set of guidelines that the company follows when developing and using AI in its CRM systems. These principles are based on the following five values: responsible, accountable, transparent, empowering, and inclusive.

50
Q

Cloud Kicks wants to implement AI features on its Salesforce Platform but has concerns about potential ethical and privacy challenges. What should they consider doing to minimize potential AI bias?

A. Implement Salesforce’s Trusted AI Principles.
B. Integrate AI models that auto-correct biased data.
C. Use demographic data to identify minority groups.

A

Answer - A

Salesforce’s Trusted AI Principles are designed to guide ethical and responsible AI implementation, including addressing and mitigating bias in AI systems. Following these principles helps ensure that AI is used in a way that minimizes potential bias and ethical concerns while promoting fairness and transparency in AI applications.

51
Q

Regarding Salesforce’s Trusted AI Principles, what is the main emphasis of the Responsibility principle?

A. Defining the technical requirements for AI integration
B. Establishing a structure for data model accuracy
C. Guaranteeing ethical AI usage

A

Answer - C

The core emphasis of the Responsibility principle within Salesforce’s trusted AI principles is to secure ethical and accountable AI utilization

52
Q

What is the main focus of the Accountability principle In Salesforce’s Trusted AI Principles?

A. Taking responsibility for one’s actions toward customers, partners, and society
B. Ensuring transparency in AI-driven recommendations and predictions
C. Safeguarding fundamental human rights and protecting sensitive data

A

Answer - A

The core focus of the Accountability principle within Salesforce’s trusted AI principles is to guarantee that AI systems are responsible and their actions can be readily understood.

53
Q

What does Salesforce’s Trusted AI Principle of Transparency entail?

A. Tailoring AI features to align with particular business needs
B. Incorporating AI models into Salesforce workflows
C. Providing a clear and comprehensible explanation of AI decisions and actions

A

Answer - C

The principle of transparency in Salesforce’s trusted AI principles primarily advocates for the clear and understandable explanation of AI decisions and actions.

54
Q

Within Salesforce’s Trusted AI Principles, what is the primary goal of the Empowerment principle?

A. Enable users to address complex technical challenges using neural networks
B. Enable users of varying skill levels to create AI applications through user-friendly interfaces, without needing to write code
C. Enable users to actively participate in the expanding field of AI research and knowledge

A

Answer - B

The principle of empowerment in Salesforce’s trusted AI principles primarily aims to empower users to understand and control AI systems.

55
Q

What is the best method to safeguard customer data privacy?

A. Archive customer data on a recurring schedule.
B. Track customer data consent preferences.
C. Automatically anonymize all customer data.

A

Answer - B

By continuously tracking and respecting customer data consent preferences, organizations can ensure that they are using customer data in compliance with privacy regulations and the individual choices of their customers. This approach prioritizes transparency and consent, which are essential principles in data privacy protection.

56
Q

What step should be followed to build and apply reliable generative AI while considering Salesforce’s safety guidelines?

A. Construct appropriately sized models to minimize environmental impact
B. Maintain transparency when AI generates and independently delivers content
C. Establish safeguards to mitigate harmful content and safeguard Personally Identifiable Information (PII)

A

Answer - C

“Establish safeguards to mitigate harmful content and safeguard Personally Identifiable Information (PII)” is the correct answer because it aligns with the principles of responsible and ethical AI development, as well as data protection.

57
Q

Which statement best reflects Salesforce’s commitment to honesty in training AI models?

A. Manage bias, toxicity, and harmful content by implementing integrated guardrails and guidance
B. Guarantee proper consent and transparency when employing AI-generated responses
C. Reduce the AI model’s environmental impact and carbon footprint during training

A

Answer - B

Ensuring that users are aware of and have given their consent for the use of AI-generated responses demonstrates transparency and honesty in AI interactions. It respects user preferences and privacy.

58
Q

What constitutes an instance of ethical debt?

A. Breaching data privacy regulations and neglecting fine payments
B. Introducing an AI feature after identifying a detrimental bias
C. Postponing the release of an AI product to retrain a data model

A

Answer - B

Ethical debt refers to situations where ethical concerns or issues are recognized but not immediately addressed or corrected. In this case, launching an AI feature despite knowing it has a harmful bias creates ethical debt because the issue of bias has been acknowledged but not rectified. This can lead to negative consequences and ethical dilemmas down the line.

59
Q

What data does Salesforce automatically remove from Marketing Cloud Einstein engagement model training to reduce bias and ethical risks?

A. Demographic
B. Geographic
C. Cryptographic

A

Answer - A

Demographic data includes information related to characteristics such as age, gender, race, ethnicity, and other personal attributes. Excluding demographic data helps prevent the AI model from learning biases associated with these attributes and promotes fairness and non-discrimination in AI-driven processes. Salesforce’s practice of excluding demographic data aligns with ethical considerations in AI to avoid bias and promote equity.

60
Q

What are some of the ethical challenges associated with AI development?

A. Implicit transparency of AI systems, which makes it easy for users to understand and trust their decisions
B. Potential for human bias in machine learning algorithms and the lack of transparency in AI decision-making processes
C. Inherent neutrality of AI systems, which eliminates any potential for human bias in decision-making

A

Answer - B

The ethical challenges in AI development primarily revolve around the potential for human bias in machine learning algorithms and the need for transparency in AI decision-making processes. These challenges highlight the importance of addressing biases and promoting transparency to ensure responsible and ethical AI development.

61
Q

Salesforce defines bias as using a person’s immutable traits to classify them or market to them. Which potentially sensitive attribute is an example of an immutable trait?

A. Financial status
B. Nickname
C. Email address

A

Answer - A

Financial status is an example of an immutable trait, which is a characteristic that cannot be changed or is highly resistant to change over time. Financial status typically includes attributes like income level, wealth, or financial stability, which are not easily altered by an individual. Using such immutable traits for classification or marketing purposes can be sensitive and potentially discriminatory, which aligns with Salesforce’s definition of bias.

62
Q

A customer using Einstein Prediction Builder is confused about why a certain prediction was made. Following Salesforce’s Trusted AI Principle of Transparency, which customer information should be accessible on the Salesforce Platform?

A. A marketing article of the product that clearly outlines the product’s capabilities and features
B. An explanation of how Prediction Builder works and a link to Salesforce’s Trusted AI Principles
C. An explanation of the prediction’s rationale and a model card that describes how the model was developed

A

Answer - C

Transparency principle - Customers should comprehend the reasoning behind each AI-generated recommendation and prediction. This involves offering comprehensive details such as model cards.

63
Q

How can Cloudy Computing enhance its AI practices while adhering to Salesforce’s Trusted AI Principles?

A. Conducting internal surveys among the company’s employees
B. Embracing AI practices in line with industry trends and competition
C. Soliciting independent feedback from external ethics experts, customers, and advisory boards

A

Answer - C

The Accountable principle underscores taking responsibility for one’s actions towards stakeholders and actively seeking external input for ongoing enhancements.

64
Q

The technical team at Cloudy Computing is evaluating the efficiency of their AI development procedures. Which well-established Salesforce model should guide the creation of reliable AI solutions?

A. Ethical AI Process Maturity Model
B. Ethical AI Prediction Maturity Model
C. Ethical AI Practice Maturity Model

A

Answer - C

This model is designed to guide the development of AI solutions in an ethically responsible manner, emphasizing best practices, transparency, and compliance with ethical principles. It provides a framework for evaluating and improving the ethical maturity of AI practices within an organization, making it the most suitable choice for Cloudy Computing’s evaluation of their AI development processes.

65
Q

Which Salesforce Trusted AI Principle highlights the significance of designing AI models to reduce bias for everyone potentially affected?

A. Transparency
B. Inclusiveness
C. Accountability

A

Answer - B

This principle underscores the need to consider and include diverse perspectives, demographics, and user groups when developing AI solutions to promote fairness and equitable outcomes. It aligns with the goal of reducing bias and ensuring that AI benefits a broad and inclusive audience.

66
Q

When should the use of natural language processing (NLP) for automated customer service be disclosed to the customer, following Salesforce’s Trusted AI Principles?

A. After they have finished their interaction with AI
B. When they specifically ask for a live agent
C. At the outset of their conversation with AI

A

Answer - C

Disclosing the use of NLP for automated customer service at the beginning of the conversation ensures transparency and informs the customer that they are interacting with an AI system rather than a human agent. This upfront disclosure promotes trust and transparency in the customer-agent interaction, allowing customers to make informed decisions about their engagement with the AI system.

67
Q

In what way should a financial institution adhere to Salesforce’s Trusted AI Principle of Transparency when executing a campaign for preapproved credit cards?

A. Clarify how risk factors like credit score may influence customer eligibility
B. Highlight sensitive variables and their stand-ins to avoid biased lending practices
C. Integrate customer input into the ongoing training of the model

A

Answer - A

Transparency involves providing clear and understandable explanations of how AI-driven decisions are made. In the context of a financial institution’s campaign for preapproved credit cards, explaining to customers how risk factors like a credit score can impact their eligibility is essential for transparency. It ensures that customers clearly understand the criteria used to determine eligibility and fosters trust in the decision-making process. This transparency also helps customers make informed choices about applying for a pre-approved credit card.

68
Q

What is one way to achieve transparency in AI?

A. Communicate AI goals and objectives with those involved prior to all interactions.
B. Establish an ethical and unbiased culture amongst those involved.
C. Allow users to give feedback regarding the inferences the AI makes about them.

A

Answer - A

Disclosing the use of NLP for automated customer service at the beginning of the conversation ensures transparency and informs the customer that they are interacting with an AI system rather than a human agent. This upfront disclosure promotes trust and transparency in the customer-agent interaction, allowing customers to make informed decisions about their engagement with the AI system.

69
Q

What is a sensitive variable that can lead to bias?

A. Country
B. Education level
C. Gender

A

Answer - C

Gender is a sensitive variable that, when not handled appropriately in data analysis or AI models, can lead to biased outcomes. It’s essential to ensure that gender-related data is treated with fairness, equity, and consideration to avoid perpetuating biases or discrimination.

70
Q

What type of bias results from data being labeled according to stereotypes?

A. Interaction
B. Societal
C. Association

A

Answer - C

Association bias, also known as associative bias, is a type of bias that arises in data when there are systematic and non-random associations between variables. This bias occurs when data labels or attributes are influenced by societal stereotypes, preconceptions, or cultural biases.

Stereotypes and Preconceptions: Association bias often results from stereotypes and preconceived notions that people hold about certain groups or categories. These stereotypes can affect how data is labeled or categorized.

71
Q

What is an implication of user consent in regard to AI data privacy?

A. AI ensures complete data privacy by automatically obtaining user consent.
B. AI operates independently of user privacy and consent.
C. AI infringes on privacy when user consent is not obtained.

A

Answer - C

User consent is a fundamental aspect of data privacy and ethics. In most cases, AI should not collect, process, or use personal data without the explicit and informed consent of the user. Failing to obtain user consent can indeed infringe on privacy rights and may lead to privacy violations or legal issues.

72
Q

What is a key benefit of effective interaction between humans and AI systems?

A. Leads to more informed and balanced decision-making
B. Reduces the need for human involvement
C. Alerts humans to the presence of biased data

A

Answer - A

Effective collaboration between humans and AI systems involves leveraging the strengths of each, particularly in the context of Salesforce’s suite of products humans and AI to work together, leveraging the strengths of each to make more informed decisions. This is evident in the design and implementation of Salesforce’s Einstein Bots, which are designed to work in tandem with human agents, not replace them.