Jess Morley
@jessRmorley
Followers
6,454
Following
931
Media
704
Statuses
7,744
💯 health data nerd. PhD @oiioxford . Previously @BennettOxford . Postdoc @DECyale
Yale, New Haven, CT, USA
Joined April 2015
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
Eagles
• 230970 Tweets
Dick Cheney
• 192590 Tweets
Packers
• 130317 Tweets
ENGFA ACTRESS 100M
• 118238 Tweets
श्री गणेश
• 113163 Tweets
Ganesh Chaturthi
• 103865 Tweets
Ganesh Chaturthi
• 103865 Tweets
गणपति बप्पा
• 81210 Tweets
Saquon
• 76054 Tweets
Jalen
• 58603 Tweets
Barkley
• 44004 Tweets
भगवान गणेश
• 41183 Tweets
#二度と撮れない画像を貼れ
• 34726 Tweets
Jordan Love
• 32308 Tweets
#GanpatiBappaMorya
• 28779 Tweets
Party Love AndaLookkaew
• 26724 Tweets
JustinOn NCAA100Kickoff
• 25401 Tweets
ビッグラン
• 15478 Tweets
イコラブ
• 12766 Tweets
#学マスデビュー初心_名古屋
• 11232 Tweets
Nikocado Avocado
• 10555 Tweets
Pinned Tweet
That’s a wrap.
Thanks
@UniofOxford
for my BA, MSc, MA, and now FINALLY PhD.
Thanks to everyone who got me here.
Thanks
@ExeterCollegeOx
for being my home from home.
And especially huge thanks to my glorious family for making today so special.
14
5
189
I was predicted BCB at A-level. I got AAB. I technically needed AAA to meet my Oxford conditions. St. Anne's let me come anyway. They understood I was more than a letter on a piece of paper. Do you know what doesn't have that kind of understanding: an algorithm.
8
71
793
⏰NEW PAPER⏰ WHERE DOES NHS DATA GO?? This is a BIG one. When
@bengoldacre
and I wrote the Goldacre Review last year, we recommended that all NHS data flows be mapped. Now Dr. Joe Zhang & I (with some help) have done just that: Some key takeaways 🧵
21
227
649
🚨ANNOUNCEMENT🚨
Today we (
@bengoldacre
& me) are launching the Goldacre review: “Better, Broader, Safer: Using Health Data for Research and Analysis.” Available: in both HTML and pdf format.
13
149
442
It’s me, hi, I HAVE NEWS.
2.5 wks ago I passed my
#PhD
viva with no corrections thanks to examiners
@pearsekeane
&
@adam_mahdi
.
So, after ~10 years in Oxford, I’m moving across the🌊 to
@DECYale
() to join
@floridi
& co.
How it started vs How it ended.
46
12
265
It’s only been a few weeks, but I’m SO TIRED of the idea that LLMs are tech’s gift to medicine. There’s more to good medicine than parroting back answers (not to mention LLMs are unregulated & generative). A 🧵 thread of my fave papers about why medicine is more than Q&A:
14
84
228
🤓FINISHED🚨 The entire
@bengoldacre
Review is now in HTML, pdf, exec sum, summ, full text & Twitter versions (to which I’ve I’ve added my personal chapter reading lists 👇). I’ll now return to my PhD on the totally unrelated topic “designing an algorithmically enhanced NHS” 😂
12
59
216
As
@NHSX
comes 2 life 2 make tech better in the NHS we want to make sure that we ONLY support tech that is ethical & responsible. This is why we have published the updated version of the Code of Conduct 4 Data-Driven HC Tech
@IndraJoshi10
Thread: 1/4
10
113
190
*ATTENTION PLEASE*
#AI
#digitalhealth
#tweeps
Today
@IndraJoshi10
and I at
@NHSX
are launching our new "Artificial Intelligence: How to get it right report" with help from A LOT of people across the digital health ecosystem 1/ 14
6
82
150
WE WON! Best paper in
#AI
4 social good category at
@NeurIPSConf
#NeurIPS2019
so supercool, so unexpected. Me appearing in the photo via phone is v “onlife” appropriate 😉 😂. So proud 2 work with
#dreamteam
@oxfordethicslab
@Floridi
&
@DigiCatapult
@libbykinsey
@anat_elhalal
1/
11
24
142
HOW CAN WE MAKE AI WORK IN THE NHS?
HOW CAN WE MAKE NHS INFO INFRASTRUCTURE FIT FOR THE 21ST CENTURY?
These are *BIG* questions.
Now you can read my (long) answer because MY THESIS “Designing an algorithmically enhanced NHS” is available online: 🧵
5
45
127
🚨PSA🚨 My (Hopefully BS Challenging) Guide to Thinking Critically About AI in Healthcare is now complete and available here: It takes the form of a themed annotated bibliography with brief summaries for each theme.A quick look at the themes 👇🏻:
*Structured Procrastination* Cannot cope with the amount of AI BS currently washing around so putting together a structured guide to "Thinking Critically about AI for Healthcare" as a means of combatting some nonsense & planning next two chapters of PhD. Shareable when done.
4
2
39
2
43
117
IT HAPPENED: "MSc Social Science of the Internet: Distinction." Didn't know those words could still make me so happy this year, but: 🥳. Esp happy that thesis rant abt (lack of) health apps efficacy & needed governance models got distinction cos now I'm sure I can publish it 😅🙌
10
2
110
I. Am. Tired NHS data are a fantastic resource that are already, and should be increasingly, used to produce benefit for *patients*, public health, social care, and the NHS. But repeating old misinformed narratives won't help achieve that goal 🧵.
12
45
112
IT IS DONE! Degree 2/3 complete. Thanks to everyone who has let me waffle on about thesis 4 wks. Thanks to
@UniofOxford
for having me for a second time (photo: 2012). Thanks to
@wellcometrust
& doctoral studentship for letting me come back to complete the hat trick in October
9
1
111
Screaming, Crying, Throwing Up: 3 years, one pandemic, OpenSAFELY, Goldacre Review, 900+ references, and 92,000 words later and the thesis is *submitted* 💀
19
1
111
So, the
#SpringBudget
included £3.4bn investment in capital to fund new technology for the
#NHS
to enable efficiency gains. Great, but let's hope it's spent wisely. Here's where I'd put my money ... 🧵
14
48
110
RIGHT. You’ve all had a weekend to digest the
@bengoldacre
review. There’s a lot to take in, so let’s do some deeper dives into content. Let’s start at the very beginning, since I’ve heard it’s a very good place to start: NHS service analytics.
🚨ANNOUNCEMENT🚨
Today we (
@bengoldacre
& me) are launching the Goldacre review: “Better, Broader, Safer: Using Health Data for Research and Analysis.” Available: in both HTML and pdf format.
13
149
442
8
30
106
2day 12yrs after my 1st
@UniofOxford
acceptance letter, I got my 3rd (DPhil). Mode of delivery has changed 4rm snailmail to email but the feeling is the same. Big thanks to
@IndraJoshi10
@pearsekeane
& esp
@Floridi
4 the support. Sorry
@oxfordethicslab
seems ur stuck w/me 🤷🏻♀️.
22
2
104
🚨PSA🚨As ALL the data strategies etc. have published at once, I've summarised 9 of the most important docs, analysing & theming the recs/commitments to provide a high-level overview of the combined strategic direction, goals & aims for NHS data.
4
28
99
It’s
#LondonTechWeek
so before we all get carried away with the hope & hype of AI for healthcare let’s take a quick dive into the challenges of actually implementing AI solutions. These (partially) explain why most AI solutions remain in the lab & aren't yet in the clinic.
🚨PSA🚨 My (Hopefully BS Challenging) Guide to Thinking Critically About AI in Healthcare is now complete and available here: It takes the form of a themed annotated bibliography with brief summaries for each theme.A quick look at the themes 👇🏻:
2
43
117
4
43
96
I'm so proud to see
@IndraJoshi10
rightly highlighted as one of the women leading in
#AI
in the summer edition of
#WomenInLeadership
@WILPublication
I can attest to the fact that Indra is a true leader, a brilliant colleague and, above all, a gr8 friend 💛
5
20
88
NEW PAPER ALERT from me,
@dr_c_morton
&
@bengoldacre
at
@EBMDataLab
& many many many
#NHS
data heroes who I will try to tag in this thread: "Bringing NHS Data Analysis into the 21st Century."
#NHSAnalytics
/1
10
25
87
New in
@TheLancet
4rm
@Floridi
& I We discuss the need to be ethically mindful re
#AI
4
#healthcare
& the need to ensure this !=
#ethicswashing
by developing a standard. Give it a read 👇& let us (
@oxfordethicslab
) know what u think
#AIEthics
#dataethics
"Ethical review [for
#AI
in
#healthcare
] cannot be a one-off, tick-box exercise and must be repeated regularly in a consistent way."—
@jessRmorley
@Floridi
@Turinghttps
://www.thelancet.com/journals/lancet/article/PIIS0140-6736(19)32975-7/fulltext
@TheLancet
@oiioxford
@UniofOxford
4
69
140
2
41
85
About 80% of my life is spent sending links to interesting & relevant papers I see on Twitter to myself in slack……..so that I can never read them.
10
2
79
There’s a LOT to say, but, briefly:
- consent is pointless unless it’s meaningful
- “public good” is a problematic, ill-defined & over-used phrase
- support for science & research is *conditional*
- ethics is > privacy
- privacy = a group level construct
7
22
77
Been working on a paper for a while auditing health apps available via the US/ UK App Stores. It's just been accepted for a presentation at the Yale Postdoc Symposium. A great opportunity to highlight the (shockingly) poor quality of most apps. Abstract 👇🏻 Paper coming soon.
12
20
79
NHS data are priceless: they should *not* be sold.
🚨NEW EDITORIAL🚨
In
@bmj_latest
me, Nicola Hamilton &
@Floridi
on why recent suggestions to sell
#NHS
data to fund a tech revolution are problematic & *must* be avoided. Plus an alternative option.
4
36
78
Don't mind me, just casually:
0
50
77
For anybody who wants my slides, anyone can read them here: or link to full text:
There is a risk of
#AI
widening
#healthinequalities
, particularly in the short term. And
@jessRmorley
says there's very little clinical evidence of AI's effectiveness in healthcare as yet.
#NTSummit
Watch the livestream:
0
21
26
5
25
76
#News
. 4rm 2day I’m policy lead at EBM DataLab at
@UniofOxford
I will cont 2 provide AI expert advice 2
@NHSX
w/
@IndraJoshi10
just not everyday. Still studying
@oiioxford
& researching with
@Floridi
&
@RosariaTaddeo
at
@oxfordethicslab
#datasaveslives
3
4
73
I’m sorry, I can’t talk about AI in the NHS right now.
Why?
Oh.
COS I’M AT THE ERAS TOUR
#TheErasTourLondon
7
1
72
Taking a week off at the end of August. What’s the 2020 version of an out of office? “Thank you for your email. I’m currently on my sofa, I’ll respond when I make it back to the kitchen table?”
6
5
71
All folks headed to
#Expo19NHS
be sure to check in with
@IndraJoshi10
and
@tara_donnelly1
to talk all things
#AI
and
#health
- especially the NHS AI Lab 👇🏻. We've invited along some friends doing some pretty cool things to provide case studies too (
#cutthehype
)
#digitalhealth
3
23
68
RIGHT the
#NHS
has turned 75 & the Government has published 21❗️policy documents relevant to NHS Data in the last 18months including last week's Workforce Plan. Anyone would be forgiven for having lost the (gg)plot. SO I've synthesised ALL THE DOCS:
7
21
67
Once, just once, I’d like to see the maths behind claims that AI will boost productivity by X% or deliver X in revenue. What is the baseline? How much does it cost to get it to work? What’re the sustainability implications? I’m all for exploring opps but please give me NUMBERS.
A new plan led by
@matthewclifford
will use AI to boost growth and improve our public services.
The AI Opportunities Action Plan will identify how to boost uptake of AI across the public and private sectors to increase productivity.
(1/2)
28
41
206
16
19
70
Using AI to speed up diagnosis won't increase system capacity to treat & without health economics re balancing specificity, sensitivity, generalisability we don't know cost effectiveness. Invest in tech sure, but also in staff retention & infrastructure
1
16
65
I have a joke about the efficacy of health apps, but there's no evidence that it's actually funny.
2
6
62
DO RISK STRATIFICATION ALGORITHMS WORK?
That is the question.
Here is our answer:
TL;DR: Results of real-world application of risk-prediction models are disappointing: equal weight of evidence suggesting a harmful effect as a beneficial one
🧵
9
31
65
AI has the *potential* to make healthcare safer & better .. yes... of course .... but only if the infrastructure necessary to make it work ethically, safely, & effectively receives the same amount of attention as the tech itself
1
16
65
It's
#WomenInScience
day* I'm wearing my female scientists top & I wrote about why closing the STEM gap matters + how we're (
@EBMDataLab
@OxPrimaryCare
@UniofOxford
) trying to contribute to the effort here:
*(+++ trans, enby, agender,+ other identities)
5
9
62
HUGE HUMONGOUS CONGRATULATIONS to
@NDevito1
of
@BennettOxford
& supervisors
@bengoldacre
@carlheneghan
for successfully passing his
#PhD
viva today with (very) minor corrections. Cheers Dr. DEVITO.
5
5
65
I'd like to take this opportunity to remind people that EHRs are far from perfect. An enormous amount of data curating is required, & deep understanding of the concepts represented in e.g., SNOMED codes, before they can be 'safely' used for research etc.
8
13
62
Not a major update, but because it's important for transparency/CoI purposes, I am happy to share that I've joined the National Data Guardian panel (see here: ). Minutes from the panel meetings are shared publicly:
4
5
63
There's a better way forward.
Fix basic infrastructure, more data in fewer places with more consistent access mechanisms, greater central oversight & more public involvement.
Invest in data quality
Review IG rules in the round to fit the new infrastructure
Tackle IP
End🧵
6
16
60
There’s domaine dressing and then there’s being in the library dressed as the library. I think I may have reached my sartorial peak.
5
0
61
Alright
@NHSX
at
#CogX19
@IndraJoshi10
& I are out. Thanks
@ChannelSwimSue
@techUK
4 hosting our final
#doubleact
event of the week despite my 🤧. On to the
#AISummit
2moro for
#AI4SG
w
@JoshCowls
@anat_elhalal
@maria_axente
to rep
@NHSX
but also
@oxfordethicslab
then 😴😴😴
6
5
57
Headed to the
@NHSrCommunity
conference to talk all things
@bengoldacre
Review & RAP.
“This is a story all abt how the NHS got flipped, turned upside down. & I’d like to take a minute, just sit right there, I’ll tell you all about how data became core to healthcare.”
2
6
54
Become an academic they said. Specialise in data policy they said. It'll be fun they said.
4wks of UK policy:
NHS data & UK digital strategies
Consultation results re UK data bill
Digital & Data roadmap
Data storage & processing infra paper
RAP strategy
*cries in overwhelm*
8
2
60
In a parallel universe where COVID never happened, and I didn’t get quarantined on Saturday, I graduated on Saturday instead! Fun to get this in the post this morning. Also I find it highly appropriate that there is a computer error on this .....
12
0
58
Today is an auspicious day:
- It's 32 yrs since I arrived on the planet
- 2 yrs since we first went into lockdown & started building
@OpenSAFELY
in earnest
- Our 1st official day as
@BennettOxford
with
@bengoldacre
as Prof:
- My dress has pockets
12
5
59
A key aim of ours in writing the
@bengoldacre
review was to challenge the false dichotomy that you can EITHER protect privacy OR you can have broad access to data for lifesaving research. IRL you can have privacy AND access. This sums it up brilliantly
1
20
58
I am aware that this is quite long and thus rather a significant undertaking to read. I am in the process of turning it into individual papers BUT in the meantime, if I recorded the lecture version of it - would that be useful/interesting?
HOW CAN WE MAKE AI WORK IN THE NHS?
HOW CAN WE MAKE NHS INFO INFRASTRUCTURE FIT FOR THE 21ST CENTURY?
These are *BIG* questions.
Now you can read my (long) answer because MY THESIS “Designing an algorithmically enhanced NHS” is available online: 🧵
5
45
127
13
4
57
📜NEW PAPER🚨 Following from our papers on the What to How of AI ethics & Ethics As A Service, our 3rd paper on 'operationalising AI ethics: barriers, enables, and next steps' is now published in AI & Society (
#openaccess
)
#AIEthics
#AcademicTwitter
🧵 1/n
3
13
58
🚨NEW EDITORIAL 🚨
me &
@drjzhn
. Why is the NHS Federated Data Platform controversial? The idea is sound. NHS digital architecture needs updating. NHS retains 'control.' The design is secure. BUT qs remain re: trust, VFM, scope, & vendor lock-in.
2
28
57
Ok, so you need to look under the hood, but a couple of standouts in the
#KingsSpeech
- electronic recording births/deaths
- info standards for health IT suppliers
- revision of digital economy act
- broader consent 4 interdisciplinary research
- stronger cybersecurity regs
4
19
55
RIGHT. Several ppl have asked & today seems a good day to discuss what GOOD NHS information infrastructure looks like & what SHOULD be designed to enable the successful use of advanced analytics inc AI👀. So let's do this, summary of thesis here:
It’s me, hi, I HAVE NEWS.
2.5 wks ago I passed my
#PhD
viva with no corrections thanks to examiners
@pearsekeane
&
@adam_mahdi
.
So, after ~10 years in Oxford, I’m moving across the🌊 to
@DECYale
() to join
@floridi
& co.
How it started vs How it ended.
46
12
265
0
7
51
You’ll see a lot on here, over the next couple of days, about AI being the “solution” to the NHS’s productivity crisis. Last week I wrote a short blog for the
@NuffieldTrust
explaining why such messages are overly simplistic.
There are continued hopes that
#AI
will help solve some of the NHS’s problems.
@jessRmorley
says it is important that the optimism for the role of AI in the
#NHS
comes with necessary realism. 👇
1
13
20
4
21
54
I try to make my work as accessible and available as possible e.g., via guide to thinking critically about Health AI (), sharing slides & my blog. But reading is sometimes annoying. I am considering making a series of 15 min videos. Would that be useful?
14
2
51
Great to be the awkward could vs. should person at
@DHRewired
today
#Rewired24
with esteemed colleagues
@Vin_Diwakar
@ConfedMatthew
@Dansjonston
@jhoeksma
. Adrenaline finally coming down. If anyone would like my sparkly slides:
#Rewired24
@Vin_Diwakar
kicks off our Digital Transformation Stage keynote with
@jessRmorley
@ConfedMatthew
@Dansjohnston
@Imprivata
@jhoeksma
0
2
14
13
5
52
For the 2nd week running NEW PAPER ALERT. This one has been a long time coming. “The ethics of AI in healthcare: A Mapping Review” with
@caiocvm
@chrisdburr
@JoshCowls
@IndraJoshi10
@RosariaTaddeo
@Floridi
We map all the epistemic, normative & overarching ethical issues at 5 LoA
3
17
52
Re-listening to His Dark Materials (I want to remember all details b4 I watch the series) & honestly the fact that Dr. Mary Malone leaves Oxford & HER ENTIRE WORLD to join a cross-cosmos war because that seems preferable to completing an academic funding app is v.relatable
0
4
50
Draft
#NHS
Data Strategy is out today . Wondering how this dovetails with our
@bengoldacre
review? ToRs are in Annex C, we’re mostly focused on the themes in chapters 3 & 5, & we’re approx 1+ month away. In the meantime, interested to hear views …..
1
18
51
Congratulations and welcome to everyone
#GoingToOxford
. I arrived at 19 and have rarely left the
@bodleianlibs
since. You’re going to love it.
Congratulations to everyone who is
#GoingToOxford
!
We are proud of you and can't wait for you to start your Oxford journey.
#ResultsDay
#ALevels2023
🎬 | Tommaso Villani
39
110
806
1
2
50
Great, but please don’t do this by forcing arbitrary performance metrics on a system on its knees or promising AI is the solution.
Start by giving GPs & practices the respect they deserve & haven’t had from policy in years.
Tackle the cause of the scramble, not the symptom.
My Labour government will end the 8am scramble for GP appointments.
4K
873
9K
4
9
48
Ever wondered why we (& others) bang on about the importance of curation when working with NHS EHR data? Ever thought, surely it’s easy to work out who has eg diabetes? Then please enjoy this wonderful insight into the brilliance & madness of SNOMED codes
3
8
46
2day's announcement of the
#NHS
#AI
Lab is the culmination
lot's of🤔abt making responsible
#AI4health
happen. We'll focus on evidence & standards (as
@IndraJoshi10
& I outline in here ) & bring 2gther academics, ethicists, technicians, clinicians, policy
Today we can announce that we're going to set up an NHS AI Lab - in partnership with our pals at the Accelerated Access Collaborative.
We're going to focus on practical AI tools that will make a difference to patients, and we're going to embed data ethics in all of our work.
38
261
755
8
14
46
Digitising the “red book” would be a win for labour & the NHS. It’s a well defined problem to fix. But let’s not pretend it’s a new initiative, rather than a new attempt. The starting point should be a postmortem understanding why previous attempts failed to avoid repetition.
2
6
46
Please let's stop publishing stories like this: "the tool detected 66% of long stayers ...a reduction of single day can lead to £1.7m savings." HOW DO YOU KNOW THAT?!? We need proper validation & quantification of impact. AI exceptionalism is bad for all.
5
8
44
FIRST DAY of (virtual) “induction week” aka FIRST DAY OF PhD!! 😬🍾🤓. I have an appropriate mug (funnier now personal/work life happen inside my laptop) & Dino face coverings: I am ready. If u r an actual fresher like me in 2009 & not a fake 1 like me in 2020
#WelcomeToOxford
2
0
43
🚨ANNOUNCEMENT🚨 Today
@NHSTransform
has published the final version of its data strategy: Data Saves Lives - Reshaping health and social care with data . It’s moved on leaps & bounds since the draft last year AND it responds to our
@bengoldacre
review. A🧵
1/4 The Government’s strategy - Data Saves Lives: Reshaping Health and Social Care with Data, has been published today by
@sajidjavid
. It sets out
@DHSCgovuk
’s ambitious plans to use data to modernise the NHS and transform health & social care.
2
13
22
1
17
45
🚨HELP WANTED🚨
Health Data/AI Tweeps I am in need of your assistance.
I'm recruiting interview participants for my PhD research “Designing an algorithmically-enhanced NHS.”
You can book an interview slot here:
But first more details 👇🏻
5
37
45
Another issue with *that*
#fatima
ad is the implied OR. We are obsessed with binaries. Did we 4get 'AND' exists? Nobody is one-dimensional. Where are the pirouetting pen-testers? The bboying devs? I ❤️ dance. Last I checked I could still boogie & write code. Both bring me joy.
2
4
44
There is something very meta about trying to get ethics approval in order to do research about ethics.
4
1
42
I absolutely cannot put finger to keyboard until I have a plan of what to write, otherwise i feel very very lost. So it’s a *HUGE* relief that ~ 600 ~ references & hours later, I know where I’m going & this giant plan exists outside of my head.
#phdlife
#phdproblems
4
1
44
Man in WHSmith’s searching for the best “write fast” pens 4 his A-level-taking daughter, asks assistant who cannot help. Me: 30+ yrs of stationary enthusiasm & more hours of exam writing than I can count - my time has come.
4
2
44
New continent, new city, flat, new university, new job.... same Jess.
Some things never change: I'll always do my best thinking in a library.
4
2
45
2 YEARS OF OPENSAFELY - A THREAD
I know it’s Friday. I know it’s sunny. But tomorrow is the 2-year anniversary since we - led by
@bengoldacre
and
@sebbacon
- launched
@OpenSAFELY
so I thought I’d take a moment to share some of its features & outputs.
1
12
43
RIGHT. Our
@BennettOxford
webinar providing an overview of current UK data policy developments, by topic, and by theme, is done!! Thanks to ALL who joined
@bengoldacre
@NDevito1
& me. The magic of
@tomodwyer
means my long-form text version is now up:
3
13
40
Today, 3 wks into Postdoc life, I received leave to supplicate 4rm
@UniofOxford
@oiioxford
@ExeterCollegeOx
, formally marking the end of my PhD & *officially* making me Dr*. Morley. Now, I can FINALLY use this sticker, & I can upload my thesis for all to read. Happy Weekend!
2
1
43
YET AGAIN there's a big-name journal article doing the rounds waxing lyrical about LLMs for healthcare cos it answers *some* questions correctly *some* of the time about *some* people. Do celebrate tech successes but don't get carried away. Be critical:
3
8
42
After a morning of talking about AI hype vs hope: A reminder that my guide to "thinking critically about AI in healthcare" or (in my words) being "skeptically optimistic" is on google drive here: . It's designed to help cut through the BS in this space.
3
21
42
"Generative AI will be transformative for the NHS" yes, BUT it'll only be transformative in the *right* way if we design it to be: I'd still like to see more investment in/ investigation of GenAI uses for backend operational efficiency not just clinical.
6
10
41
⁉️What should the next gov do to make AI work in the NHS?.” ⁉️
As we approach LDN data week & the GE, I’ve been asked this many times
The v.long answer is in my thesis.
BUT few folk want to read 90k words so here’s an 80 min video podcast version:
1
14
42
🌟 NEW PREPRINT 🌟 from
@oxfordethicslab
"Ethics of Algorithms: Key Problems & Solutions" covering normative, epistemic and traceability issues w/ Andreas,
@nikitaggarwal
@JoshCowls
@HuwJRoberts1
@RosariaTaddeo
@Floridi
THREAD /1
2
16
40
Interested in China 🇨🇳,
#AI
,
#policy
and
#ethics
? Check out our
#OA
@oxfordethicslab
paper on “The Chinese Approach to AI: An analysis of Policy and Regulation.” 1st author is the brill Huw Roberts w/ input 4rm me
@JoshCowls
@RosariaTaddeo
and PI
@Floridi
#happyreading
Open Access article on "The Chinese Approach to Artificial Intelligence: An Analysis of Policy and Regulation", comprehensive (covers Chinese sources) and updated, by Huw, Cowls, Morley, Taddeo, Wang, Floridi,
@oxfordethicslab
@oiioxford
1
24
89
2
13
42
Admittedly the bar is not very high, but please meet the best thing to come out of 2020: FINCH! Not a replacement for our gorgeous Scout (2010-2020) but a wonderful new addition to the family, helping to fill the hole he left. 10 points if you know how the names are connected.
8
0
39
How it started vs How it’s going: obsessing about patient data & matching my mug to my outfit since 2018.
1
0
39
👋🏻 🇬🇧: see ya later London
It’s me, hi, I HAVE NEWS.
2.5 wks ago I passed my
#PhD
viva with no corrections thanks to examiners
@pearsekeane
&
@adam_mahdi
.
So, after ~10 years in Oxford, I’m moving across the🌊 to
@DECYale
() to join
@floridi
& co.
How it started vs How it ended.
46
12
265
3
0
39
Jenna Rink was right, being 30 is much better than being 13 but 2003 is not looking so bad right now! Very weird way to enter my 3rd decade
#sociallydistancing
in my flat on my own but c’est la vie - at least much time for reading! Wash your hands 4 me & enjoy living
#onlife
😉
19
0
38
*Structured Procrastination* Cannot cope with the amount of AI BS currently washing around so putting together a structured guide to "Thinking Critically about AI for Healthcare" as a means of combatting some nonsense & planning next two chapters of PhD. Shareable when done.
4
2
39
It’s almost as if we need standardised methods for evaluating efficacy and cost efficiency of software (esp AI) before we let it a) be sold direct to patients; and b) disrupt well-evidenced and thought through NHS care pathways. Who knew?!?
Babylon always claimed that their AI would eventually start to help reduce the costs of delivering remote GP consultations. This was the central fallacy.
Babylon is the most expensive failed experiment on digitising primary care to date.
3
62
200
2
12
38
D2 of
#Cogx19
You can catch both of these
@turinginst
#awardwinners
on the ethics stage & health stages at 11 & 12.30 respectively. Also catch
@sanakb
@JeniT
ethics stage at 10 &
@IndraJoshi10
at 5. I’ll be
@NHSX
in the expo - come say hi before I 💯lose my voice
#AI
#HealthTech
1
9
36