TheProve.pptx

The CxO Guide to Generative
AI: Threats and Opportunities
June, 2023

Yes, It’s real!
Over the last few months, we’ve had
conversations with leadership teams
that start with a high-level question:
“Will Generative AI impact our business,
and how? The board wants us to develop
a Generative AI strategy.”
With a certain level of conﬁdence the
answer is YES,Generative AI will impact
all industries, directly or indirectly.
According to OpenAIResearch,around
80% of the U.S. workforce could have at
least 10%of their work tasks affected by
the introduction of Large Language
Models like ChatGPT,while approximately
19%of workers may see at least 50% of
their tasks enhanced.
Ifyou are a shareholder,board member
or CEO,it is imperative to develop a
Generative AIstrategy.
As of June 2023, the majority of CEOs
and business owners have already
experienced the impact of ChatGPT
in their personal lives, noticed its
effects on professional productivity,
and discussed it in strategic business
conversations.
This report is focused on practical
opportunities, risks, and limitations of
the state-of-the-art Generative AI
applications for enterprises of all sizes.
We’ll review business patterns, as well
as technical, legal, security,and
governance aspects to consider when
transforming the enterprise with
Generative AI.
©2023 Provectus. All rights reserved |provectus.com

4
10
Discover your competitive advantage:
Emerging opportunities to embrace and use cases 1
1
Start your journey now:Practical steps
for adopting Generative AI 19
Connect with
our leadership team
1
2Act now:Existential threats and emerging markets.
Will my industry and business model exist in 10years?
3
4
5 25
Background and baseline alignment:
What is actually happening and why?
Outline

Background and baseline
alignment: What is actually
happening and why?

LLMs help to overcome
writer’s block in Gmail or
Outlook
Google Search uses LLM
models to summarize
answers that appear
on the ﬁrst page
Translation and speech
recognition services are
the ﬁrst widely adopted
applications of
transformer models
Intuition behind Generative AI:Existing applications
in everyday life that you may never have noticed
While driving, the Tesla’s
autopilot is powered by
the similar architecture of
pre-trained transformers

Eliminating noise and adding clarity
generation of new chemical weapons.
Appropriate programs to address the
issue are being developed by regulators
in collaboration with major AI players.
No.Generative AI is a powerful yet practical
technology that has its own inputs, outputs,
and limitations. We suggest not to extrapolate
it to general intelligence, and to completely
eliminate this distractor from real-world
business conversations.
Yes and No.“We too” low-hanging-fruit use
cases are to be commoditized soon. Innovative
business models, however,require a much
deeper understanding the ﬁrst principles of
Generative AI and a solid data infrastructure.
No.It is recommended not to confuse
Generative AI technology with applications like
ChatGPT that are not designed for enterprise
requirements. But there are multiple vendors,
including major cloud providers, who address
or will address these issues soon.
Are humans going to be
replaced by AI in the future?
CEOs might not be aware
of all technical details to
innovate on business models
with Generative AI.
Gen AI applications are not
secure and not compliant,
so major enterprises and
even whole countries ban it.
The Distractor
“Stop AI for 6 months.”
Elon Musk, Bill Gates, and
1000+others signed a petition
to halt AI experiments.
Our Perspective
● Open-source software can’t be regulated
or stopped at its core.
● There are a lot of harmful use cases like
Itis important to align on
frequently asked questions
and common distractors.
Doing so will help businesses
leaders eliminate noise and
stay focused on achieving
real business outcomes.

Itdidn't happen overnight.
Two major forces have
converged and pushed the
technological breakthrough
that we are all experiencing
today:
Background: Why the market
shift is happening right now
Advancements in deep learning, generative
research, foundation models over the last
15years paved the path for existing
state-of-the-art model architectures.
“Attention is all you need” –a breakthrough
paper from 2017 is considered as a starting
point of the most recent innovation cycle. It
took the industry 5 years to build an actual user
facing application (ChatGPT) and develop a
commercial model for selling it to end users.
Data storage, Cloud and GPU compute
became more accessible. Prior advantages of
internet giants and their almost exclusive
access to unlimited scale is no longer the case.
Small startups like OpenAI and Stability AI can
now develop new breakthrough applications
on the same infrastructure as Google or Meta.
1
2

The strategic question that must be
raised on every board agenda:
“Will our indusĞry and business model
be relevanĞ in 10 years?”
“If AI can drive a car,I’m sure iĞwill
disrupĞ my indusĞry sooner or laĞer.”
–Brij Patel, Founder and CEO of AFG
An example of an immediate
protective legal action item for
content-heavy businesses is to
update their data terms of use.
Act now!
Don’t wait until your industry is disrupted by AI!

Discover your competitive
advantage: Emerging
opportunities and use cases

Improving employee
productivity with AI
Examples:
Generative AI boosts worker productivity
by 14%in the ﬁrst real-world study.
At Provectus, we see a productivity
increase of 50-90% in tasks such as new
domain research and business analysis.
We've also seen a 40-80% productivity
boost in the generation of synthetic
datasets for ML model training and quality
assurance. Engineering tasks –from
requirement analysis and legacy code
base research to bootstrapping
new applications, testing, and debugging
–have seen a performance improvement
of 30-60%.
Monetization of your
assets through AI
Examples:
Tesla is monetizing its enhanced autopilot
and self-driving add-ons for $6k and $15k
each, respectively,which account for 50%
of the total car price. In this sense, Tesla is
becoming an AI business.
GitHub is charging $19per enterprise
user for its AI Copilot add-on, representing
a 100%price increase from the basic plan.
Bloomberg has announced
BloombergGPT, creating an opportunity to
begin monetizing its data assets through
a new interface.
Think big!
Assess two major types of opportunities

Generating new construction materials, chemical
solutions, tissues, proteins, drugs and clothes designs
For ISVs:simplifying complex user interfaces of
ERP/CRM/Accounting applications by exposing all their functions
via natural language interface.
For publishers and educational/media content owners: selling
existing proprietary content and data products through a new
conversational interface and consumption-based business model
A unified conversational interface on top of a corporate
knowledge graph, internal analytical data sources, and
public content
All manual workflows, legacy Robotics Process
Automation (RPA)tools or Business Process Outsourcing
(BPO) workflows can be dramatically simplified and
streamlined with Generative AI
All previously developed and maintained customer
support chatbots, personal assistants, lower-level NLP
models can be rebuilt or replaced by new LLM APIs
Impact
level
Impact Approach Example use cases
Opportunities by level of impact
high
low
Creating and leading new market categories
through strategic differentiation
Increasing market share and revenue by
augmenting your customer experiences
Decreasing costs by increasing workforce
productivity
Reducing costs by improving
general operational efficiency
Optimizing total cost of ownership
for legacy ML applications
Inventing completely new business
models and creating new markets
Copilot for your customers.
Inventing new customer interaction
models and modernizing legacy call
centers, web and mobile interfaces
Copilot for your employees. Empowering
your leaders and individual contributors
with new productivity tools, so that they
can deliver better quality at reduced cost
Putting your non-critical business on
autopilot by introducing an extreme level
of automation at every level with AI
Tech enterprises who have previously
invested in ML applications are to
modernize them by using a new stack
of foundation models

Hypothetical
example:
“Generate a new stainless
material that can
withstand heat of 1
,650
Kelvin and endure
multiple landing cycles of
the spacecraft”
sĞrengĞh
durabiliĞy
Ğh
ermal conducĞiviĞ
y
ﬂexibiliĞy
GenAI tech blueprint
prompĞ
Diffusion
Model
3D aĞ
omic sĞrucĞures
of new sĞ
able
maĞerials
ouĞpuĞ
DaĞabase of exisĞing
maĞ
erials and Ğh
eir
properĞies
Ğ
rained on
Moonshot use case example:
Novel materials generation
Intuition:
In a similar fashion, Stable Diffusion
models can generate new images.
The same architecture can be used
for generating new materials with
desired physical properties.
Risks:
2 3
1
Availability
of training
datasets
Upfront Feasibility
investments horizon is
into R&D 3-5 years

Media
&Entertainment
● Content search and discovery
● Content generation
(text,art,music)
● Content moderation
● Scenario development
● NPC generation in gaming
● Commentary generation for
sports events
Financial Services
&Insurance
● Client reporting and compliance
reporting
● Credit spreading
● Intelligent document processing
and understanding
● Intelligent due diligence
● Augmented client interactions
Consumer
Internet
● Horizontal and vertical search
transformation and market value
redistribution
● Augmented customer experience.
● Specialized personal AI agents
● Content licensing for AI vendors
Education
● Micro learning and AI-augmented
experience
● Educational content monetization
through conversational UX
● Educational and training content
generation and new monetization
models
Professional
Services
● Specialized productivity tools
● Internal knowledge graphs
● Conversational business
intelligence
● Client reporting generation
Healthcare
&Life Sciences
● Medical charts and clinical notes
generation and understanding
● Pricing and reimbursement
modeling
● R&D knowledge graphs
● Protein folding simulation
● Creation of new drugs, proteins
and materials, and their assessment
Opportunities by industry

Opportunity:
Client and Compliance Reporting for
financial asset management firms is a
resource-intensive task.
Solution:
Use a Generative AI-powered report builder
that simplifies and accelerates portfolio
performance analysis, narrative generation,
and the final report production.
KPIs: Risks:
● Fiduciary duty for
providing accurate
and valid information
● Time to produce and
deliver a report to the
client,leading to
increased customer
satisfaction
● Cost of producing and
maintaining client and
compliance reporting,
directly impacting the
operating margin
Input prompt
Natural language prompt:
Use case details: Client reporting in Financial Services
Generated output Context drilldown
Mitigation:
● Data lineage and cross
validation of every
metric and data point
in the report

Executive
team (CxO)
● Generative business intelligence
and solutions for tapping into company’s
data and insights and summarizing it for
the C-level format
● Augmented communication strategy:
a. Board reporting
b. All-hands updates
c. Company values
Sales &
Marketing
● Sales call assistant
● Market research and reporting
● Proposal generation
● Prospect research and due
diligence
People &
Operations
● Improving employee experience
through internal knowledge
graph and AI search
● Augmented onboarding
● Augmented screening interview
Research
&Development
● Industry speciﬁc R&D knowledge graphs
and AI-curated industry insights
● Simulation and experiments tracking
● New materials, chemicals, protein designs
Security &
Compliance
● Compliance reporting
● Security risk assessment
● Augmented security training
● Background check
Finance &
Legal
● Augmented contract review and redlining
● Vendor assessment and due diligence
● Reporting
● Scenario simulation and stress testing
Opportunities by business function

Use case details: Sales and Operations Planning
Opportunity:
Sales & Operations Planning process
requires a comprehensive visibility and
intelligence to accelerate decision making
process by command center leaders, as well
as CRO,CSCO — all up to CEO.
Solution:
Deploy a Generative AI-powered S&OP
visibility solution to provide conversational
decision intelligence on top of internal (ERP,
MES,CRM) and external (risks, news,
sanctions) knowledge graphs.
KPIs:
● Revenue through
proactive mitigation of
supply chain disruptions
Risks:
● Overall maturity of
internal ITsystems
● Operational efﬁciency of
the command center
● Adoption and change
management of
legacy business
processes
Identify risks and mitigation strategies using natural language

A Generative AI
adoption framework.

Generative AIAdoption
Journey at Enterprise
Off-the-Shelf LLM Enablement
Achieve productivity gains by equipping your knowledge workers with
out-of-the-box LLM applications, and establish a solid foundation for
secure, compliant,fair,and responsible AI adoption from day one.
Enterprise AI Search
Extend out-of-the-box LLM capabilities with an enterprise knowledge
base that includes, but is not limited to, FAQs, internal policies, training
materials, and information from IT,HR, Product, Sales, and Marketing
departments. Empower knowledge workers to be more productive in
their day-to-day tasks.
Generative Documents Intelligence
Boost productivity for corporate roles dealing with large, unstructured
documents in underwriting, auditing, asset management, and legal
affairs by reducing the time required for quality document processing
and understanding from days to minutes.
Generative Business Intelligence
Increase the productivity of knowledge workers by exposing them to
corporate business intelligence capabilities and KPIs through a natural
language interface, to accelerate critical business decisions.
Universal Enterprise Agents
All of the aforementioned features and components can be combined
into a comprehensive, easy-to-use product that serves as a universal
Gen AI companion for enterprise users.

How to start
a Generative AIjourney?

OPPORTUNITY
Use
Case
KPI
Expected
Impact
How strong is the
expected impact
on business KPI?
Busines
s
Proxy
KPIs
IMPACT
Tip:
No matter how excited you are
about technology,try to find a
solution for your business
problem, and not vice versa
Tip:
Challenge your use cases for
compliance and fairness
requirements early on
in the process
Tip:
You need to have a hands-on AI
Expert and AI Product Owner in
the room for a quick feasibility
assessment,and to create a
high-level ROI model
ML
Problem Integration UX/UI Costs
How difficult
is it to obtain
and prepare
data?
How
complex is
the
algorithm?
How difficult
is it to
integrate and
deploy?
How difficult
is it to
integrate the
solution into
the existing
UX/UI?
What is the
assumed
cost level?
EFFORT
Data
Step 1
:Prioritize use cases
User
s
Categor
y

business
impact
(monetary
value)
adoption risk
(human aspect)
high
low
high low
Use Case 1
:
Customer support (business function)
High impact on customer satisfaction
Low risk perceived –routine script-based jobs are already commoditized
and outsourced
Executive consideration
Introduce a Generative AI-powered chatbot for frequent routine
and FAQ-type support cases as an entry point for customer inquiries
to make human agents assist only with complex cases
Where to start
Focus on business outcomes –test drive a solution pilot
with a small subset of the audience, fine-tune, and scale.
Use case 2:
Content creation (individual contributors)
High impact on productivity
High risk perceived –a perceived threat of workforce elimination
Executive consideration
Redefine the role of content creators to shift their focus on providing
inputs and supervising the AI-generated content vs. creating the first
drafts
Where to start
Focus on change management –cultivate an AI-first culture and
build from success with early adopters within the team.
no go
no go
2
1
Use cases prioritization: Opportunity-Risk-Strategy Tool

Step 2.1:Understand Gen AI Terminology: AI, Generative
AI, GPT, Large Language Models, and Foundation Models
Artiﬁcial
Intelligence (AI)
Generative AI
provides
algoriĞhms
parĞof
Large Language
Models (Ğ
exĞs)
GPT-4,Codex, Falcon,
Claude-2
parĞof
Foundation Models Machine Learning
Ğ
ype of
Diffusion Models
(images)
Stable Diffusion
examples of FM
ChatGPT, GitHub Copilot
examples of end-user models &
Ğools
DreamBooth,
Midjourney, DALL-E 2

Step 2.2: Understand available tools and services
Fully Managed
Services
Infrastructure
&Compute
Frameworks
(ChaĞGPT plugins) (OpenAI Service) (Amazon Bedrock) (GenAI App Builder)
Open-source
LLMs
(commercial
use allowed)
Application Components
ProprieĞ
ary
LLMs
(GPT models API) (PaLM API) (Amazon TiĞ
an)
(Jurassic-2 API)
(Claude API)
and many
oĞhers…
PrompĞ
Engineering
LangChain LlamaIndex
Model
Hubs Amazon
SageMaker
JumpSĞarĞ
VecĞ
or
DBs
TS,UL2 Bloom S
Ğ
ableLM
Dolly
PyĞhia, GPT-NeoX Open AssisĞanĞ
Amazon
SageMaker Azure ML GCP VerĞ
ex AI
and
many
oĞhers…

Step 3:Understand risks and risk
mitigation strategies
Harmful responses,
hallucinations, and inherited
biases from the data that AI
has been trained on
● Design human-in-the-loop experience for continuous evaluation and data collection.
Don’t use LLM as part of a fully automated pipeline without human in the loop
● User training and clear communication about possible AI behavior
● Apply LLM alignment methods like Reinforcement Learning from Human Feedback (RLHF).
Note: it is an advanced technique mastered by a handful of LM vendors so far.
● Turn hallucinations into a positive benefit for creative use cases like music and art generation.
Environment impact from a very
intensive process of carbon emissions
during LLM training and inference
● Budget carbon impact in advance when selecting an LLM vendor
and include it into the total cost of the model
● Stanford AI Index report is one of the latest data sources
for estimating an environmental impact of LLMs
LLMs can’t be used for
precise calculations,
structured finance, and
mathematical problems
● Multiple deterministic systems (databases & APIs) can be integrated together
and be exposed through an AI interface
Compliance risks of data leakage
when sending it to third-party
providers
of foundation models
● Private VPC deployment of your own model, as well as 3rd-party models, is available for private
deployment
● Implement logging and audit trail for capturing, filtering, and blocking of any sensitive requests
to AI models
Knowledge cut-off ● Integration with real-time data sources and APIs, and with internal knowledge graphs
Data copyright and data licensing
● Implementing data lineage, data cleansing, and data testing best practices
● New legal frameworks and processes to be implemented
Cost of LLM hosting
● Mix large & expensive models for creative tasks and your own “smaller” models for specialized tasks
● Right now it’s a dawn for open-source LLMs, datasets, and optimization techniques
GenAI Risks &Limitations Mitigation Strategy

Decision hub: AI Center of Excellence
Is it a core part of our business model
and strategic differentiation?
Investment timeline: Months or years?
Do we have an R&D budget?
Do we have an expertise?
What are our compliance requirements?
What are our privacy requirements?
Level of commoditization. Should it be
available as a service?
What is our governance model?
Use
case
Build (train)
Buy
Subscribe
Customize
(ﬁne tune)
Acquire
Step 4: Deﬁne your implementation strategy

Connect with our leadership team
Oleg Blokhin
Executive General Manager
Strategic Accounts
oblokhin@provectus.com
Stepan Pushkarev
Co-Founder
CTO and CEO
spushkarev@provectus.com
Matt B. Ewalt
Executive General
Manager US West
mewalt@provectus.com
Yogesh Joshee
Executive General
Manager EMEA
yjoshee@provectus.com
Dave Zabihaylo
Executive General
Manager US Central
dzabihaylo@provectus.com
Rinat Gareev
Principal
AI Solutions Architect
rgareev@provectus.com

Questions?
We will be happy
to answer!
provectus.com
125University Avenue
Suite 295, Palo Alto
California, 94301
hello@provectus.com

TheProve.pptx

More Related Content

Similar to TheProve.pptx

Recently uploaded

TheProve.pptx