Optimize AI at Scale With Platform Engineering for MLOps
Just as platform engineering emerged from the DevOps movement to streamline app
development workflows, so too must platform engineering streamline the workflows
of MLOps. To achieve this, one must first recognize the fundamental differences
between DevOps and MLOps. Only then can one produce an effective platform
engineering solution for ML engineers. To enable AI at scale, enterprises must
commit to developing, deploying and maintaining platform engineering solutions
that are purpose-built for MLOps. Whether due to data governance requirements or
practical concerns about moving vast volumes of data over significant
geographical distances, MLOps at scale require enterprises to utilize a
spoke-and-wheel approach. Model development and training occurs centrally,
trained models are distributed to edge locations for fine-tuning on local data,
and fined-tuned models are deployed close to where end users interact with them
and the AI applications they leverage. ... Enterprises should hire engineers
with MLOps experience to fill platform engineering roles appropriately.
According to research from the World Economic Forum, AI is projected to create
around 97 million new jobs by 2025.
The Blockchain Integrity Act: Latest Attempt to Restrict Financial Privacy
In short, the Blockchain Integrity Act would first establish a two‐year
moratorium that prohibits financial institutions from going anywhere near
cryptocurrency that has been routed through a mixer. With that two‐year
moratorium in place, the Blockchain Integrity Act would then require the
Department of the Treasury to study how people use mixers and other
privacy‐enhancing technology. ... The second half of the legislation—the
request for a study—is less concerning if it’s considered alone and without the
surrounding context. The request seeks information regarding different types of
privacy‐enhancing technology, illicit and legitimate use history, and an
analysis of what the government’s role might be here. Those are all reasonable
inquiries. Again, without additional context, it’s an encouraging sign that
Representative Casten is interested in learning more about how this technology
is used for both better and worse. Yet what isn’t encouraging is that
Representative Casten introduced the bill saying that “until we’ve studied
[privacy enhancing technologies like mixers] and have a good audit trail, the
presumption should be that these are money laundering channels.”
Some strategies for CISOs freaked out by the specter of federal indictments
“Some CISOs feel like they’re the frog that’s in the water that’s starting to
boil, and they don’t like that feeling, and they want to make sure that
they’re doing the right things to navigate that heat,” Sullivan said during a
panel discussion, “CISOs Under Indictment: Case Studies, Lessons Learned, and
What’s Next,” at this year’s RSA Conference. The panel of current and former
CISOs emphasized that in this environment, CISOs need to document their roles
and responsibilities, involve the right people in incident response and
decision-making processes, and have the courage to stand up for their
convictions to minimize the risk that they will face the same fates as
Sullivan and Brown. ... “The heat is up because the reality is you’ve got
these entities in government who are responding to a huge rise in cybercrime
in a way that no one can hide. It’s not like in the old days when if an
incident happened, most people wouldn’t notice when stuff happens. Today, the
whole world notices,” he said. Blauner’s bottom-line advice to CISOs to
protect themselves is to “take a look at every governance document you’ve got
and really make sure that it’s crystal clear about roles and responsibilities,
especially around who makes risk management decisions.”
Wearable devices can now harvest our brain data. Australia needs urgent privacy reforms
In a background paper published earlier this year, the Australian Human Rights
Commission identified several risks to human rights that neurotechnology may
pose, including rights to privacy and non-discrimination. Legal scholars,
policymakers, lawmakers and the public need to pay serious attention to the
issue. The extent to which tech companies can harvest cognitive and neural
data is particularly concerning when that data comes from children. This is
because children fall outside of the protection provided by Australia’s
privacy legislation, as it doesn’t specify an age when a person can make their
own privacy decisions. The government and relevant industry associations
should conduct a candid inquiry to investigate the extent to which
neurotechnology companies collect and retain this data from children in
Australia. The private data collected through such devices is also
increasingly fed into AI algorithms, raising additional concerns. These
algorithms rely on machine learning, which can manipulate datasets in ways
unlikely to align with any consent given by a user.
Cloud environments beyond the Big Three
The resurgence and innovation in edge computing and on-premises technology
further support the trend toward diversification as data generation and
consumption locations continue to spread geographically. ... Edge computing
addresses these limitations by processing data closer to where it is
generated. This drastically reduces latency and enhances the user experience
in applications such as IoT, retail tech, and smart manufacturing. Although
many consider edge computing to be small devices, it also includes entire data
centers and smaller server installations that exist to serve a specific
business location. Many enterprises don’t see the wisdom of sending their data
on a 2,000-mile round trip to the point of presence for a public cloud
provider, which happens more often than we understand. Additionally, although
the cloud offers good scalability and flexibility, concerns over data
sovereignty and security continue to push certain industries towards
on-premises solutions. Sensitive data and critical applications in sectors
such as finance, government, and healthcare often necessitate keeping data
in-house under strict regulatory frameworks.
Controlling chaos using edge computing hardware: Digital twin models promise advances in computing
Using machine learning tools to create a digital twin (a virtual copy) of an
electronic circuit that exhibits chaotic behavior, researchers found that they
were successful at predicting how it would behave and at using that
information to control it. Many everyday devices, like thermostats and cruise
control, utilize linear controllers—which use simple rules to direct a system
to a desired value. Thermostats, for example, employ such rules to determine
how much to heat or cool a space based on the difference between the current
and desired temperatures. Yet because of how straightforward these algorithms
are, they struggle to control systems that display complex behavior, like
chaos. As a result, advanced devices like self-driving cars and aircraft often
rely on machine learning-based controllers, which use intricate networks to
learn the optimal control algorithm needed to operate efficiently. However,
these algorithms have significant drawbacks, the most demanding of which is
that they can be extremely challenging and computationally expensive to
implement.
Digital recreations of dead people need urgent regulation, AI ethicists say
Such services, which are already technically possible to create and legally
permissible, could let users upload their conversations with dead relatives to
“bring grandma back to life” in the form of a chatbot, researchers from the
University of Cambridge suggest. They may be marketed at parents with terminal
diseases who want to leave something behind for their child to interact with,
or simply sold to still-healthy people who want to catalogue their entire life
and create an interactive legacy. But in each case, unscrupulous companies and
thoughtless business practices could cause lasting psychological harm and
fundamentally disrespect the rights of the deceased, the paper argues. “Rapid
advancements in generative AI mean that nearly anyone with internet access and
some basic knowhow can revive a deceased loved one,” said Dr Katarzyna
Nowaczyk-Basińska, one of the study’s co-authors at Cambridge’s Leverhulme
centre for the future of intelligence (LCFI). “This area of AI is an ethical
minefield. It’s important to prioritise the dignity of the deceased, and
ensure that this isn’t encroached on by financial motives of digital afterlife
services, for example.”
How To Take The A-I-M Approach To Leadership
I like to break down the concept of taking aim into three components, which I
call the A-I-M approach: appreciation, imagination and motivation. The common
thread across all three of these principles is communication—and leaders
cannot be effective without it. Showing genuine gratitude is a foundational
aspect of effective leadership. Expressing heartfelt encouragement
demonstrates empathy and humility. And this simple show of appreciation
directly benefits the organization by motivating employees to continue
contributing to the company’s success and nurturing their loyalty. ... A
leader’s job is not to be the author of all ideas but to inspire team members
to tap into their imaginations and present fresh approaches to solving
problems, delivering solutions and communicating with clients. ... One
of the responsibilities of a leader is to understand what moves their teams
into action. As author and leadership coach John Maxwell famously wrote, “A
leader is great not because of his or her power, but because of his or her
ability to empower others.” I call that motivation.
Colorado AI legislation further complicates compliance equation
CIOs might struggle with the bill’s language because the focus is on whether
AI — in any form — helps make “consequential decisions” that could impact
Colorado residents. The bill defines consequential decision as being any
decision “that has a material legal or similarly significant effect on the
provision or denial to any consumer,” which includes educational enrollment,
employment or employment opportunity, financial or lending service, healthcare
services, housing, insurance, or a legal service. ... Another provision could
prove onerous for CIOs who do not have full knowledge of every AI
implementation in use in their environment, as it requires companies to make
“a publicly available statement summarizing the types of high-risk systems
that the deployer currently deploys, how the deployer manages any known or
reasonably foreseeable risks of algorithmic discrimination that may arise from
deployment of each of these high-risk systems and the nature, source, and
extent of the information collected and used.” ... One especially dicey area
in the legislation that should concern CIOs is when AI — especially generative
AI — acts on its own.
AI's Game-Changing Role in Finance and Audit Processes
Auditors can face several risks when using AI. These risks include
over-reliance on AI-generated insights, potential biases and quality issues
from incomplete or poor-quality data andcybersecurity threats such as
consequences in terms of hacking of the confidential data from the AI
websites. Thus, it is necessary to ensure compliance and implement
safeguarding measures. Following are some of the possible measures that can be
implemented to mitigate the above-mentioned risks. Human judgement: While AI
is a great tool to be incorporated in the professional world to help auditors
and organisations streamline their existing processes, AI work on standard
algorithm that can’t be customised on case-to-case basis. Therefore, to ensure
the accuracy of the results, a human review can be placed in practice to
review from and validate the accuracy of output results. Updating back-end
algorithms: The better the algorithms, the better the results. Regular updates
to the back-end algorithms can yield more accurate and improved outputs,
adapting to changing scenarios and data formats, ultimately mitigating the
risk of incorrect or inaccurate results..
Quote for the day:
"Don't find fault, find a remedy." --
Henry Ford
No comments:
Post a Comment